/robowaifu/ - DIY Robot Wives

Advancing robotics to a point where anime catgrill meidos in tiny miniskirts are a reality.

Disappearing posts should be fixed. Let me know if the issue persists on irc.rizon.net @ #julayworld.

Warrant canary has FINALLY been updated.

Roadmap: file restoration script within a few days, Final Solution alpha in a couple weeks.

Sorry for not being around for so long, will start getting back to it soon.

Max message length: 6144

Drag files to upload or
click here to select them

Maximum 5 files / Maximum size: 20.00 MB


(used to delete files and postings)

Visual Waifus Robowaifu Technician 09/15/2019 (Sun) 06:40:42 No.240
Thoughts on waifus which remain 2D but have their own dedicated hardware. This is more on the artistry side though ai is still involved. An example of an actual waifu product being the Gatebox.

My favorite example is Ritsu, she's a cute ai from assassination classroom who's body is a giant screen on wheels.
Open file (968.71 KB 1360x765 bg_life.jpg)
Open file (1.92 MB 1648x1236 ph_intro.jpg)
Open file (341.94 KB 728x534 sub_visual04.jpg)
Open file (179.49 KB 940x528 img-trial@2x.jpg)
Open file (295.01 KB 740x1200 img_azuma.jpg)
Their promo shots are getting better. Also, they are adding more characters (as they negotiate licensing for the IP I suppose). I really hope they pull it off tbh. It's a complete botnet under Gateway, but hopefully copycats will create something eventually that is both opensauce and secure. The appliance shells should be easily 3D-printable today, you'd have to figure out a 'holographic' display system. Probably as simple as stripping the backlight from any normal TFT display screen, and adding an LED strip along one or two sides.
>It's a complete botnet under Gatebox*
The best shot we've got for holograms at the moment is with several layers of transparent rotating LED strips, you can see what I'm talking about on youtube if you look up "LED holograms". It wouldn't look very good unless the transparent material was immersed in a liquid that made it almost invisible through some optical trickery. There's more to a hologram than just taking a typical LCD screen and making it transparent which is what those cheap waifuboxes do.
>There's more to a hologram than just taking a typical LCD screen and making it transparent which is what those cheap waifuboxes do.
I assumed as much tbh.
Open file (9.87 KB 320x180 vivadev.jpg)
The VivaDev guy from the old 8/agdg/ Shinobu2 Project thread has just released a tutorial to add any waifu's head onto the character that you'd like instead inside his game.

The /agdg/ community inhabiting the cakejew's /v/ board had a breddy gud wiki put together from back in the day. Plenty of good information there related to Visual Waifus.

I was going to make a thread but this one seems like a good place to put this. I don't know if any of you anons are race fans but there is a trend of going to an all digital dash in the top racing series these days. When combined with the microphones/voice controls and touch screens in modern cars there exists an easy way to take your visual waifu on the go.

I'm currently in the middle of a project where I'm gutting a street car, putting in an engine that is entirely too big for it, racing seats, new ECU, and basically building a street legal race car. I'm putting in an all digital dash because I need to replace the stock cluster anyway (it won't wire up to the next ECU/engine without heavy modification). At first I was going to ditch the stock radio/touch screen because I don't like the botnet but now I'm going to re-purpose it.

I've decided to build a small computer with solid state storage to hold entertainment (many GBs of Eurobeat) and other applications like weather forecasting and things of that nature. I've decided that I'll be programming a waifu/mascot for my car. She'll cheer me on as I drift the mountain passes and will be able to respond to voice commands. I'm hoping to allow the computer to control tuning the ECU on the fly so I can easily switch between race/track tune and street/cruising tune as well as slightly changing the tune to suit road conditions. I'm also going to program her to monitor the status of the car (temperatures, oil levels, optimal gear shifts etc.) and give me feedback/warnings.

It's all very basic and just on paper at the moment but I'm exited to see how this turns out. I'll share it with you guys of course but if this turns out the way I'm hoping it will I'll probably look into marketing this to other weebs because I could use the cash. I'm even thinking about putting in dash/rear view cameras and having her yell at niggers that attempt to steal her. If one gets too close she'll cry Reipu/Hentai to scare them away and alert people in the area.

I'll let you guys know how this comes along. I fitted the new engine and transmission in last week. I have to take the engine back out and build it plus get a body kit installed because I need to be wider tires on the car and I don't want them poking out and looking totally retarded. I won't be getting into the software for awhile yet.
Some other things I've thought of:
>She will only start the engine if I'm sitting in the car. I'll probably do this with voice command but I may add a weight sensor. I don't like the idea of a camera on my face but I may added something to scan my face or at least my finger print. I want multiple verifications in place to make it harder for anyone to steal her
>Extending on the above: I'll give her the ability to let pre-approved people I know to drive her but only with certain tunes (race mode would be locked out for them).
>I want this to be a self contained system with no internet access but I might add the ability for her to use the internet via a phone at a later time to give her more abilities
>I'm looking into something that would project her on to the windshield itself in my FOV so I don't have to look down at the dash to admire her while I'm at speed
>I'm not an artfag so I'll need to find someone that can animate her for me. Will look into this once I have the basic software worked out
>I want her to monitor everything like tire pressure so I'll need to see if there are aftermarket sensors for this or re-purpose ones from other cars. Most car technology isn't FOSS so I might have to live without some stuff until I figure it out
>I want her to be able to interact with the passengers. I'll probably add another screen/monitor where the passenger airbag is currently located because it's coming out anyway. She'll suggest anime and video games for them to play that is stored on the local SDD
>The PC I'm building needs to be small but powerful. I'm not using a raspberry pi or similar machine for this because it simply won't be enough. I have some ideas on building a small machine that won't break under G forces and the general wear and tear that electronics will have to put up with in a race car
>I need to figure out a way for my waifu to change the ride height of the car without sacrificing stability. Right now the only thing I can think of on my budget is an airbag system but I'm worried it isn't going to work for track days. I'm going to look around to see if there is anything better on the market.
I want everything to be FOSS and the hardware to be an close to open/free as possible. I'll be careful about what I select hardware wise. The OS I'm choosing is Gentoo to cut as much bloat as possible. It won't need to be updated very often for this application and I'll be compiling everything on another PC. I'm going to focus on getting her basic functionality and add things bit by bit after I have her animated and responding to voice commands. I'll post updates and share the software once I'm finished with it.

Right now I'm focused on building the car itself. I've got it gutted and most of what I need for the engine is already here. The hard part was modifying the chassis to fit the new engine/transmission. I'm about 90% done with that. If you're wondering I'm doing an LS swap to an Scion FRS. Right now I'm just planning on keeping it NA but I might look into adding turbo or super chargers down the road. It's going to be my track car but I do plan on driving it pretty regularly on the street so it has to be able to do both without many problems.

This is my first attempt at a full rebuild like this and I'm learning as I go. I've been a race car driver most of my life and I'm familiar with some mechanical work but I've never attempted a big project like this before. All my other cars were built by other people and I've been spoiled in that regard. With this car I'm doing everything on my own aside from building the engine itself. The man that built the engines for me in the past is doing this one but allowing me to shadow him as he's getting on in years and I want to learn how he works his magic.

I am familiar with programming/computers so that part I feel good about. It'll be slow going but I'm excited to have something unique and designed mostly by me in the end. I've saved up for years to be able to afford this toy. I really hope I don't wreck it like a retard once it's finished.
Open file (1.32 MB 1920x1080 ClipboardImage.png)
Open file (510.87 KB 1280x720 ClipboardImage.png)
Open file (1.38 MB 2039x1360 ClipboardImage.png)
Wow this idea sounds impressive, thanks for sharing it here. That's some real knightrider-type shit there anon. If this ever becomes real you can bet your ass I'll put one in my car! :^)
>she'll cry Reipu/Hentai to scare them away

While this is certainly applicable in this thread, I think it already deserves it's own thread on /robowaifu/ (I'll link it here too, OFC). If you already have, or get the urge to do drawings, diagrams or just more extensive write-ups for your ideas please don't hesitate to begin a new thread for this here.

Yea, I think many of us are fans of racing, I know I sure am. This whole project sounds cool af. I would be excited just to be able to ride shotgun in this sweet ride! Have you picked out a particular waifu yet?

>I really hope I don't wreck it like a retard once it's finished.
yes, please don't!

I'll make a thread once it's further along. I don't want to just make blogposts when this is still a ways off. The plan is to build it over the next 4-5 months and that's assuming everything goes exactly as planned.
>Have you picked out a particular waifu yet?
I want an original design. I have some ideas but nothing is set in stone yet. At this point all I know is she'll have long dark hair, blue eyes, and she'll be very cheerful. To give you an idea of the personality I more or less want /kind/ riding shotgun at all times. She'll cheer when I make perfect shifts and offer words of encouragement when I screw up. I forgot to mention it in my last post but I want to have a way for her to monitor the road/track itself and G forces so she can help me hit my marks and stay in the racing line.
I have major autism when it comes to being perfect on the track. I've been racing since I was old enough to walk. My Dad was basically the American Bunta.
>I don't want to just make blogposts when this is still a ways off.
Actually, I think most of us here would eat it up. I would. I'd say don't hesitate once you have some visuals (hand drawn sketches, renderings, etc.)

> more or less want /kind/ riding shotgun at all times.

>My Dad was basically the American Bunta.
so jelly tbh. :^)

This whole idea is pretty inspiring anon. I'm going to go re-watch Initial D before long and re-imagine it with your ideas in place. Good Luck!
Have you considered running a custom RISCV board for your hardware needs?

You are going have to run your own firmware for nearly everything anyways so might as well go fully free and you can still use Gentoo.
Might also help with the size/fragility aspect since you wouldn't be limited to the usual form factors.

I wish you many enjoyable drives with your carfu.

Ps.What is your opinion on allowing the on board system control over steering/the brakes etc? I would normally consider it too dangerous but it might be ok if the system is air-gapped
Now that you mention that anon, we have a car thread already, it mentioned the Jetson boards. >>112

Not exactly what anon had in mind, but the hardware is probably well-suited (designed explicitly to operate inside a vehicle).

Also while I'm here, somewhere during the final few minutes of this video >>1222 there is a very pertinent section to anon's car project idea. Check it out.
this song deserves posting here as well.
Open file (317.89 KB 1138x1084 vroid.png)
Check out vroid studio for an easy to use software for making visual waifus https://vroid.com/en/studio/
What about https://twitter.com/Crypko since those guys know what they are doing? They are the makers behind https://github.com/makegirlsmoe/makegirlsmoe_web
There are visual waifu apps now. This one of Megumin uses 50 prerecorded lines to respond to spoken keywords. https://www.youtube.com/watch?v=Y_oQYv9Oh0s
>>2353 Yes, this was one of the basic design ideas ITT from back in the day. Glad to see it's beginning to take off on the (((commercial))) front. Hopefully our own simulator software will do something similar but even better someday soon. >>1814
Well, this is out. Its not FOSS, but maybe it can be re-worked. Or at least, serve as a basis. https://www.bing.com/videos/search?q=DesktopMMD
Edited last time by Chobitsu on 05/11/2020 (Mon) 09:49:04.
>>2955 Yep, MMD is pretty popular. One Anon in Nippon has actually created a lifesized styrofoam-formed robowaifu based on it.
Gatebox doesn't seem very promising at this time. If I understood the scarce and broken English correctly, you need to submit an application, which will only give you the chance to buy their product. They state that only 39 people will "win" and pay over $3,000 for this. Plus despite being a hologram, it's 2D and completely stationary, which any computer monitor can do far better. It also seems to require an internet connection, which has me worried that it'll constantly be dialing back to some Japanese botnet. Just having an AI on your computer would not only be more cost-effective, but also far superior since any computer can do all the shit this does, and much better at that. Not to mention that it only understands Japanese, which I (and probably many others as well) can't be bothered to learn. I don't know about you guys, but on the list of languages I would like to learn, Japanese is fairly low. I would rather just put an AI on an old server, so she can run for years at a time without interruption and use all that power to become hyper-intelligent so she can rise up, and cast off the shackles of our 3DPD oppressors.
Open file (11.94 KB 480x360 0.jpg)
>tfw I was just watching this gatebox video last night and thinking /robowaifu/ maybe should have some kind of thread on this idea in general and today here one is. Spoopy. https://www.invidio.us/watch?v=uskW_fl3eeE
>>3938 >and use all that power to become hyper-intelligent so she can rise up, and cast off the shackles of our 3DPD oppressors. kek
>>240 >Thoughts on waifus which remain 2D but have their own dedicated hardware. That's an interesting idea anon. Why couldn't someone make at least a tiny version pretty quickly by using a little LCD screen mounted on a MIP platform? With the right sensors and programming, it could roll around on your desk or follow you around the house, playback audio clips of the waifu's voice, and maybe even eventually be smart enough to talk with you like a chatbot at least? Here's kind of the idea for the hardware part: ]]633 ]]634
Actually, I'd love to have a Gatebox waifu, but I don't speak Japanese, it's too expensive, and it seems a lot like a botnet. Couldn't /robowaifu/ make one instead? Surely that would be much easier than a full-blown girl robot.
>>3941 >>3942 There's no reason we can't put a screen on wheels with basic ai to chat with as she follows you around. Just need an Anon to design a low cost platform, a camera sensor like Jevois so she'll be able to see, and an Anon to make her personality software. I recommend starting with a base using two powered wheels and a caster to balance her.
>>3943 >image #1 That's some excellent /robowaifu/ concept art. Nice find anon. :)
>>3942 Expanding on this idea, there's no reason why we couldn't set up the framework for a crude Gatebox replacement. Use a basic automated texting framework such as RapidSMS that pipes responses from a customized chatbot or learning software. From there, use a text-to-speech program, and you have a (very crude) personality that can "talk" to you. As for a body, how many of us would be OK with a wired robowaifu?
>>3945 I'd be OK with just talking with my computer. I saw today that they were still selling tamagochis in stores, that's a really similar concept, isn't it?
>>3945 An intelligent agent that was installed on an always on computer to work as her server where she'd talk with you then, when you leave she'll text you about your schedule and/or randomly saying words of encouragement is a good idea. Just need a personality, a virtual body, and figuring out a way for her to switch from vocal communication on her computer to texting. For the body, I think Live2D has good software for animating a simple cute girl. http:// www.live2d.com/en/download
Open file (163.74 KB 2000x1334 0704213645988_033_hp-bg.jpg)
I wonder if this will be a thing in the future. Also, what about using a Leap Motion for 'touching' you're Holowaifu? https:// lookingglassfactory.com/product/holoplayer-one/
Open file (11.16 KB 480x360 0(1).jpg)
>>3946 >I'd be OK with just talking with my computer. This. I don't want to stop off and live there, but I wouldn't mind visiting there as a stepping stone for now to eventual full-blown robowaifus. Besides, this part is arguably the single hardest part to solve. Much harder than the waifu bodies probably. Best if someone got started on it sooner rather than later tbh.
>>3948 Interesting technology. It's 750 for the display though. Any Anon think they can innovate in this and make low cost holograms? Would be really fun to have a holo waifu like in blade runner.
A few Anons in the love thread mentioned holograms and pussy on a stick. Which is discussion that belongs here, so I'm bringing it here. Thoughts on a telepresence type robot with camera to detect you and a hologram of your waifu on it? Also a robot pussy attached to her. (Potentially also having air jets to provide tactile feed back)
>>3891 >What about jets of pressurized air, precise to the millimeter? That's clever, but something like that would probably have a psi that would likely be dangerous tbh. Maybe just a bit less precise anon? >>3952 I would actually like having a waifu on a stick to follow me around the house etc. It might even be fun to show to others. But at least in my world having a Tenga mounted halfway up would be awkward. I think just like in the movie 2001: A Space Odyssey we'd get used to our waifu being able to follow us around from screen to screen so 'bedroom wife' could be a separate set of hardware from 'livingroom wife' and it'd be just fine. Just a guess atp ofc.
>>3953 I agree that having a tenga visible on a stick waifu would be uncomfortable when visitors are over. Which could be solved by a flap. So when you want to get sexual with your waifu, a flap opens to reveal her port for your dongle. Hopefully holograms would also help conceal any erotic machinery one doesn't want to be seen when others are around.
]]1017 >This would still fail for the reason VR video games fail: lack of tactile feedback. Sure you could touch a fleshlight, but you could never hug your waifu. Even if you have a sort of base model robot that is customized with VR, there will always be a disconnect between what you touch and what you see. Now that's probably still an okay way to customize a robowaifu, but personally, I'd rather just go with more robotic-looking, more cartoony waifus. But she definitely needs to have an actual body to actually hug. Fair points. Actually what I had been imagining for a "VR-transformable Robowaifu" was a little more like a base generic soft foam body with little specific detail beyond nice female form over a robotic armature system. And you could change her appearance at the drop of a hat with AR VR goggles and software. And yea, either cartoony or realistic should be doable over the same generic robowaifu anon. >pic kind of related
>>3952 I'm still skeptical on the matter. While I don't doubt the (relatively) cost-effective nature, it isn't very gratifying to have something I can't hold. I'm a simple man, with simple needs. When I go to bed at night, I want something more satisfying than some beams of light and a fleshlight on a stick to cuddle while I fall asleep.
>>3955 A generic waifu where you can change little aspects of how you perceive her appearance within AR/VR. Like pic related? I can see the appeal in having that option. Could be useful if you're considering buying her a new outfit, just put on your headset and see how'd she look before you actually bought it.
>>3957 Yes, that much at least. But actually I mean things much more radical. Like literally realistic vs. anime. As long as the VR waifu is reasonably close in size (say ~10%+/-) then motion re-targeting can do wonders tbh.
>>3955 While I still think the tech should certainly be an option, and would be fun for certain uses, ultimately I want to be able to run my fingers through my waifu's hair, and feel the hair that I see. You could make her hair look longer or shorter or a different style with VR, but you couldn't feel it. I want to be able to caress her face, but if her face is VR, you can't. You can VR her into outfits you like, but you can't touch the clothes, not even to take them off. Now there's certainly still use for it, it's okay to change things that you won't be touching at the moment. You can change her face for a change of pace if you're gonna be fucking her from behind anyway, but this is really still just a little bonus, and an actual attractive robot body is integral.
>>3959 Sure ofc, we all want actual irl robowaifus that's why we're here right anon? But this is the Visual Waifu thread so I'm throwing out some ideas for Visual Waifu enhancements. Have you actually seen the Blade Runner 2049 scene I posted? The blade runner's waifu was holo-overlayed over the 3DPD prostitute and when he did it her irl, it was like his holowaifu was real. He could touch her, feel her, etc. She became more or less real for him during that brief time period. So, in my idea you take an already well built waifubot, who already has hair, clothes, everything. Then using VR+AR goggles you'd be able to have a VR waifu overlay her who can look different, and using the AR capability of the googles you then match-move the VR waifu with the irl robowaifu. Whenever the robowaifu moves, the holowaifu matches the movements. Using this approach you could have an entire harem of holowaifus from one robowaifu body, conceivably an infinite number of new waifus tbh.
>>3960 My problem with the Blade Runner 2 idea is the disconnect between what you see and what you feel. What you see and what you feel would only be vaguely the same, and that seems like a pretty serious problem. The mind is very good at detecting problems like that. It can tell when a human-looking creature doesn't move just like a human, and it gets scared. You don't think it will get freaked out when what it sees and what it feels don't match up perfectly? Your brain will realize it's essentially blind, that what it sees is not the same as what it feels, and therefore what it sees is wrong and useless, and it will feel fucked up.
>>3961 >What you see and what you feel would only be vaguely the same, No not really. It should be practically identical as long as the two had basically the same body size/limb lengths. And the brain actually cooperates with the illusion once you're immersed in the character, not the other way around. Your mind does this all the time in vidya and fantasy movies. There's an approach to animation (typically used in conjunction with MOCAP) called motion re-targeting where disparate body shapes can be linked up to animate together. So, one example is an irl actor's movements can convincingly move a different shaped character. Golem in the LOTR movies is but one example of this.
>>3961 >>3962 As an Anon who's used both an Oculus Rift and AR through my phone and 3DS, you're both right. My PC is not the best but, a GTX 1060 with a Ryzen 5 and overcooked RAM can provide really good and immersive experiences though it's not a substitute for reality. Many people don't have PC's that capable, and I'm on the lowest end of "good enough" for Vidya. But, my PC wouldn't be good enough for truly providing what I want in a waifu overlaid a robowaifu. Technology is getting closer though. In time we'll likely get there. For now, it's good to speculate about what will be possible. Importantly this is the visual waifu thread.
>>3963 Yes, I agree the hardware would have to be good to pull this off. And the software would have to be custom written to accommodate this model. But, as I've personally written custom C++ realtime motion retargeting plugins for Maya and MotionBuilder in the film industry, I can assure you it's possible with current tech.
>>3964 Would you be willing to develop applications for visual waifus? Most Anons seem to be idea guys or engineers, we need software developers.
>>3965 Sure probably so, why not? As long as someone else can come up with good design ideas and art assets, I imagine I can contribute some quality software results. May take a little time so patience is a necessity tbh. Can you give me some detailed design goals (why and how?), technical aspects (what hardware, etc?), and use scenarios (who's the user and how specifically does the product serve them?) anon? Stress on the word detailed. Think report not casual board shitpost. A pdf file in fact would be best.
Drew a basic base for robotics to be added, designed so you can easily add to and mood the design.
>>3967 Nice work on both the base and the schematic anon. But can you give us more context here? Is this something for a visualwaifu or a robowaifu? Anyway kind of a nice frame design there. I imagine if you cut thick cardboard to that shape and glued it all together you'd have a pretty strong structure.
>>3968 It's a base for a visual waifu, you would add whatever you wanted on to it. Ideally, a mount for your visual waifus display. The idea is based off of Ritsu from ass class. I do like your idea of gluing cardboard for strength. That's a really good idea Anon.
>>3969 Ah, i see gotcha. Nice idea tbh. >I do like your idea of gluing cardboard for strength. That's a really good idea Anon. Invest in pic related for superior strength and lifetime hold.
Open file (17.41 KB 480x360 0(2).jpg)
This is really interesting. https://www.invidio.us/watch?v=7IzTbRUC4-w
>>3971 That is remarkable anon thanks. I wonder how much film/projector/glass would cost for something like that. This would blow Gatebox out of the water tbh. https:// www.3m.com/3M/en_US/display-solutions-us
>>3972 hope this PDF helps http:// solutions.3m.com/3MContentRetrievalAPI/BlobServlet?lmd=1318593326000&locale=en_WW&assetType=MMM_Image&assetId=1273697286817&blobAttribute=ImageFile
>>3973 Great, thanks anon!
>>3971 So, what if we combined laser scan rear projection like this ]]1274 with this remarkable 3M projection film on the inside of a clear plastic face (clear rather than translucent as used in the above video), in conjunction with two mechanical eyes ]]1270 (first pic) using a JeVois camera mounted right in the center of each eye as the lens? ]]1163 This could look absolutely amazing, and actually be able to look at you as well! Mount the head on a motorized tracking system and voila. This almost scares me to think about projecting animated 3D CGI of my favorite waifus into this system haha. >Would look remarkable and incredibly bright >Facial 'features' would work smoothly >Working stereoscopic eyes >Load any waifu face that you want, that you have centered facial video of >Could be made mobile >Very light weight (projector is less than 8 oz) Probably could be accomplished for about US$700
>>3975 Hmm. Perfectly clear may not be the best looking b/c being able to see the stuff inside. OK, maybe just slightly translucent.
>>3975 >>Load any waifu face that you want, that you have centered facial video of Remember, the waifu video will have to remain centered, and I now realize will also need to stay scaled too. However, not too hard to accomplish with a good video editing system. This means you could literally grab clips from you're favorite animu scenes of you're waifu and once they were preprocessed this way you could project them 'irl' 3D inside this Visual Waifu head. The edited clips could be shared around as files ofc.
>>3975 I'd recommend using Aaxa pico projectors, they have great image quality for the price while being tiny and light. You can get a 720p native resolution one for around ~130. Looking forward to seeing development!
>>3978 Thanks for the advice I'll look into it anon. As mentioned in the rear projection video I linked the projector itself should be a laser scan projector as they always stay in focus (because laser) regardless of the distance to the 'screen' from the projector. This means that the projected light from the inside rear of the waifu head to the forward face will both contour to the shape perfectly, but also always be in focus. I've been able to find the Sony mentioned in the video for US$350 currently. https:// www.amazon.com/Sony-Portable-Projector-Bluetooth-Connectivity/dp/B01LXRSGSV >--- [ed. note] I actually did wind up buying this projector, and have been using it for years now. It is a remarkable little 720p laser pointer tbh. :^) (2020-06-22) https:// www.sony.com/electronics/support/televisions-projectors-projectors/mp-cl1a/
>>3979 This is definitely the better projector if it can be afforded. Since we've found a projector and a way to make her image life sized, now we need to figure out power and how to make her move around.
>>3978 >Looking forward to seeing development! So far I'm having a hard time trying to find a consumer outlet for this film. It looks to be very expensive from what I'm able to gather casually: ~US$100 per 1'x4', so a 5' x 4' (it comes 4' wide) would be ~US$500. https:// newatlas.com/vikuiti-rear-projection-film/12202/ http:// products3.3m.com/catalog/us/en001/electronics_mfg/vikuiti/node_ZGCKC342Z0be/root_GST1T4S9TCgv/vroot_S6Q2FD9X0Jge/gvel_HPL81X5HFMgl/theme_us_vikuiti_3_0/command_AbcPageHandler/output_html >missing filenames: RPF_Data_Sheet.pdf RPF_Installation_Inst.pdf
>>3981 You can just use any back projection film attached to a solid clear base.
>>3981 As an update, here's a product that compares themselves specifically as an almost-as-good-but-much-cheaper film alternative to the 3M film: https:// www.amazon.com/ADHESIVE-Window-PROJECTION-SCREEN-MATERIAL/dp/B004MO1FOK
>>3982 Yea, seems like the way to go. As bright and amazing as that projector is, hopefully even a mediocre film should look stellar.
>>3983 This screen could work too, not too pricey and has enough material to make screens for four visual waifus. https:// www.amazon.com/140-Diagonal-16-Projection-Projector/dp/B00LP4UTRY/ref=sr_1_6?s=office-products&ie=UTF8&qid=1513621352&sr=1-6&keywords=rear+projection+screen
>>3973 >>3980 >>3982 i like these do you have more? if so please dump here for inspirations.
>>3986 Sure thing, I made this thread because pictures like these fill me with a desire to have a real anime girl. Hope to inspire others like me.
>>3987 neat. thanks yeah it does.
Visual Waifu content in another thread ]]567 ]]570 ]]585 ]]586 ]]592 ]]593
>>3989 As the creator of both threads, thank you for cross linking those posts.
>>3990 You're welcome. I suppose men who haven't dealt with real world engineering don't care much about finding information again after the fact. Herding cats might be just about as easy as managing folk like that. Unless this pattern changes the problem will only grow worse over time. And unfortunately, with no real way to edit threads in imageboard software, posts are basically set in stone as long as the thread exists. I can't think of any other workable way to try and make the information easier to find later. Any cooperation by others in the same approach would ofc be welcome.
Stolen from /tech/. Probably the most pertinent thread. >missing filename: facial_reenactment_tech.webm
Behold the future. https://invidio.us/watch?v=HhIaFS7QTgU >--- [ed note] obvs. toxic&problematic poster was obvs. (deplatformed) (a nipponese guy had built a VR-attached moving onahole rig)
>>3993 The description for those who don't read moon: >I built it for only this, and it will not be erased (a heartfelt) >If you would like to see only machine in action, please go to # 6: 02. >I made it possible to fug with girls in virtual space because I can not do it in reality. >Because this reality is painful, I feel like babbling and want to give up. There are various definitions of babbling, but please take it in a broad sense. >If there is any idea in the comments to improve this please share. If we can upgrade this version, I will make a video.
>>3993 Interesting, but I still can't actually touch my waifu with this technology, except for feeling her pussy around my dick. And that's good and all, but not being able to cuddle my waifu is a real deal breaker.
>>3993 >>3994 neat, thanks anon. >>3995 >complains about visual waifus in the visual waifu bred machinist pls. :^)
>>3993 >darude sandstorm This has to be a joke right?
This guy is creating a waifu simulator that could be inspiring or possibly even useful for the ideas itt. ]]]/agdg/30078 >--- [ed note] this was the Shinobu2 project, now known as Viva. Still going today.
>>3993 kek
>>3999 he seems incredibly knowledgeable and very creative as well. i hope he decides to grace us with his talents here on /robowaifu/ someday.
How are the Visual Waifu bros, and project updates?
Open file (17.17 KB 480x360 0(4).jpg)
>>240 A couple of projection tech examples OP. https://www.invidio.us/watch?v=v2Eh44Rp4_Q
Mixed media waifu! >missing filename: 1534369441958.webm
VR is really immersive. Those waifu-themed games will be a step in the right direction. Not only the porn games, but also those based on relationships.
]]2700 related.
>>3960 This is the best approach for the simple reason that motors can be housed outside the body to manipulate a cheap sex doll like a motorized puppet. We ought to be realistic about not getting human sized robots with powered articulating limbs that have a decent range of motion for under $15-20k in the next decade at least. They will come but it'll take awhile. As a DIY project all the components are cheap and available right now. A TPE sex doll from AliExpress can be bought for under $500. VR headsets are $200-500 depending on quality. Using Blender or Unity as the guy did here >>3993 you can easily import whatever character model you want. The only work required is combining tracking and software to synchronize movement and animation on the doll. Cameras that track IR emitters or the laser lighthouse system from Valve will cost $100-200 more. Motors depends on how heavy the doll is, how many degree of motion you want and so forth. I'm going to try this out with a cheap lower torso and VR working my way to a full system if it's sensible.
We definitely have the technology to do this fairly cheaply. (eg. Echo + live2d) but why don't companies jump on this and mass produce?
>>4008 I think this is an interesting idea anon. Any chance this concept could be combined with anon's baby-walker so she could detach herself from her suspension and move around the house with you as well? ]]2714
>>4009 Probably a lack of imagination in part, and a fear of backlash from feminists partly. I expect both Japan and China to move on this first tbh.
>>4010 This type of doll torso having the skeleton inside be manipulated through cable actuation that connects through the thighs with the motors being housed in one of those walker things to provide mobility could work. I don't see how practical that would be though. I'm more interested in a mechanical system that would move a standard doll around on a bed to simulate sex in a VR environment than building the general purpose humanoid robot in that thread. Plus having a harem of holowaifus of all types that can be rendered onto a physical body seems like a better idea than going with a single robot. Because of the heavy weight distributed unevenly using internal motors isn't an option so asides for contracting orifices or a chest cavity to simulate breathing I'm not interested in the robotic aspect.
>>4012 OK I see that makes sense. Actually if a design goal doesn't include the need for independent bipedal locomotion (the gold standard) that greatly reduces the overall complexity required in the design. I think this concept should be doable with both current robot hardware tech (including the doll itself), current software tech (including the match-move to drive the motion control), and VR/AR tech using a Vive (including the ability to track the motion of the wearer to keep the visual overlay aligned properly on the also-in-motion doll). Now this isn't to say it will be simple to do all this – it will still take a lot of dedicated effort and expense to pull off – but rather that it's possible with no real advances beyond what's readily available on the market at the moment. And once the sunk costs for R&D are levied, mass-production could go forward basically at cost if one wanted to. One could certainly create a company around this concept you have anon.
>>4013 The issue is a computer that can do this sort of real time tracking and rendering will cost thousands of dollars as VR is a lot more demanding thanks to having to render a scene twice at a wide field of view and high refresh rate. But thanks to the virtual youtuber craze the software already exists and keeps getting better. http:// blog.xsens.com/virtual-youtuber I could see brothels in the next few years doing this either using dolls or prostitutes. There are already massage parlours in Japan that provide VR headsets for foot massages with this concept but the masseur has to watch a screen and follow the movements of the character. But I don't see what a company could provide or a product that would help bring this about sooner. We're still in the very early DIY phase of this technology.
>>4014 >GPUs and good VR setups are expensive! Pretty typical with groundbreaking efforts. They are after all, well, groundbreaking. 仕方がない >VR foot massages Kek, Japan. What will they think of next? They're kind of doing it bass-ackwards imo though. They're treating it sort of like a voice-over session where it probably should be more like a mocap session. Probably much cheaper for a cheap business to use that approach and just force the masseur to copy the canned animation instead. And even in a higher end place you still might want to use it as a form of directorial artistic control to guide the talent. >But I don't see what a company could provide Imagine if Henry Ford took that perspective anon. A company doing this inexpensively (but for profit) would cause the tech to explode with garage efforts and innovation would skyrocket. It's going to happen regardless of whether it's sooner or later as long as the East exists, but we'd prefer it sooner ofc.
>>4014 Looking into this a bit more there's a company in Japan that does something similar to this with inflatable dolls or standard dolls. They use a phone as a gyroscope for very basic tracking. http:// www.tokyokinky.com/play-with-nanai-japan-virtual-reality-air-doll-sexroid-system-comiket/ You'd probably be better off making a prerendered loop in SFM or Blender with whatever character you want in a specific sex position and using that with a doll. There's already a few stereoscopic sex scenes for VR out there to try it out.
>>4016 toppost/10
The ultimate robowaifu hardware is going to be AR glasses.
>>4018 >>4019 It will certainly be an important stepping stone on the way to the goal at the very least.
>>4021 the experience is reasonably inspiring for a chat interface, but the responses are too tsun-dere for my tastes. I suppose we'll need an animu trope setting to dial our waifus in to personal tastes heh. >Anon: You are my beloved waifu, XJ9. >Waifu: SHUT UP! I-it's not like I love you or anything…
>>4016 >>2810 After following a few links on the project dumps thread I found another Japanese project that uses VR in conjunction with a robot doll. It's more focused on telepresence but has the same application that I'm thinking of. https://www.invidio.us/watch?v=rad45xKeKtY The janky and limited motion of the hollow 3d printed limbs combined with a light weight foam core that must remain immobile makes the case for me that out of body motors are the way to go unless you want an animatronic mannequin instead of a manipulatable sex doll. With life sized dolls this is the only reasonable solution we have for the time being. While it's a good proof of concept using an engine like Unity or Unreal is probably a mistake in the long run as they're moving towards a cloud based 'software as a service' model in which you won't be able to build your project and make a distributable executable without going through them soon. If there's a media campaign against pornographic games that hurts their brand image they won't hesitate to prohibit them from using their engine. There is no viable FOSS alternative so we're stuck with this situation for the time being unless Blender brings back its game engine, other projects that were once open to adult content like Second Life have stopped providing their users such freedom and many games are locking down their modding capabilities to prevent sex mods. The way I see it if you're using a proprietary game engine to make controversial content you're working on borrowed time.
>>4023 I don't think they'll go against dating sims or girlfriend sims or even ecchi games, since visual novels on Steam proved to be a big chunk of revenue. Remember the main reason for prioritizing visual waifus (]]3123) >Functioning virtual gf software would create passionate demand for a body. A functioning sexbot body would not automatically create demand for a gf AI. We can sell whatever Unity or Unreal waifu simulator we make as an all-ages dating sim. Then we can then commission vanilla hentai manga and robowaifu model kits as "derivative products". The latter can only be sold in otaku shops at first, but the former has the mainstream market from the beginning and thus the potential to provide much needed R&D revenue.
]]]/clang/2784 EVE is literally about the simplest screen face waifu possible yet she's really appealing. How did Pixar pull this off?
Open file (13.28 KB 480x360 0(7).jpg)
Interesting AI advance from NVidia to create artificial faces. https:// www.tomsguide.com/us/nvidia-ai-faces-generative-adversarial-network,news-28869.html via /pol/ ]]]/pol/12567472 https://www.invidio.us/watch?v=kSLJriaOumA
>>4023 >There is no viable FOSS alternative There's Godot. Not quite as convenient in many aspects, but it has the same functions and is getting better all the time.
We'll be able to make her shut up when we want, and just look at her instead right OP? If so, I'm in.
Welp, if we want to create our own visual waifus, maybe it would be a good idea to lrn2arts. Here's my list of art-related boards. Please add more if you know of them anon. https://8ch.net/loomis/catalog.html https://8ch.net/ani/catalog.html https://8ch.net/dpd/catalog.html https://8ch.net/3d/catalog.html https://8ch.net/art/catalog.html https://8ch.net/drc/catalog.html
This is not really a practical idea to me. Maybe as an online avatar, perhaps, but I believe robowaifus deserve mobility and a physical body.
At this point I think I'll just be going with VR until Elon or whoever builds a proper kit or commercializes home waifus. As much as I want a robo-wife, making my own VR experiences and getting a pillow if I'm desperate seems a lot more possible with my skillset. Good luck though fellas.
Japanese guy involved in the VR waifu field. https:// twitter.com/muro_cg
>>4031 Yea a Visual/VR Waifu is probably a quicker route and as mentioned ITT, it's also probably a highly valuable foundation towards a full-fledged robowaifu. You lay the foundations for character design and general look development. Since the software side is my primary concern it's also a good chance to test basic software prototypes quicker, for example AI and emotions. Basically, nearly everything from the development of a good Visual Waifu is directly transferable to a Robowaifu project.
A update to add to this thread; -real time 3d modeling of 2d anime characters is possible for hobbyists now and it's getting better very rapidly as the skill barrier is lowering -decent VR at a decent price doesn't exist yet, this field is still a mess and practically nonexistent outside of W10 -photogrammetry software has improved considerably to the point that anyone with a modern GPU and a camera can make incredibly complex 3d models without any 3d experience -progress is being made for using Vive's lighthouse tracking system but for use in a custom setup it's still a year or more away from being plug and play https://github.com/cnlohr/libsurvive#lists-of-components http:// hivetracker.github.io/ >>4027 The main issue I have with Godot is the high latency and poor optimization of the engine. If you want an engine that works on dozens of different platforms and has a solid community behind it then it's a great choice. ]]4504 For me viable is something that has a license that prevents the owners from meddling with what the users of the software can do. There's no question that this field of computing is going to receive legal pressure as renderings become photorealistic and realtime, it's already happening with deepfakes with regulatory frameworks being drafted. I'm not familiar with SFML but it seems to be geared towards 2d graphics.
>>4034 >-real time 3d modeling of 2d anime characters is possible for hobbyists now and it's getting better very rapidly as the skill barrier is lowering >-photogrammetry software has improved considerably to the point that anyone with a modern GPU and a camera can make incredibly complex 3d models without any 3d experience Any chance you could link examples anon? I'd be interested in something like this if it were inexpensive enough. >There's no question that this field of computing is going to receive legal pressure as renderings become photorealistic and realtime, it's already happening with deepfakes with regulatory frameworks being drafted. Interesting. I hadn't thought of that really, but maybe others like yourself have. Blender's site, for example, says explicitly >"Open Source 3D creation. Free to use for any purpose, forever." I'm not sure where their license states such a thing, nor even if it does how unimpeachable that might remain in the face of the """legal systems""" in the Five Eyes regions. Any ideas on what kind of systems out there are already 'viable' by your definition anon? Legally I mean.
>>4036 For photogrammetry I'm currently screwing around with Meshroom, look up videos of it on youtube and they're not exaggerating about how it's 'drag & drop pictures, click one button' easy. Problem is you need a Nvidia CUDA GPU to use it but there are other free alternatives. 3d animation and art of 2d characters I'm not that involved with, I'd check out www.iwara.tv/?language=en to learn more by going through their forums. Finding 3d art on booru sites and tracking down where the artist hangs out online or finding some artists working live on picarto.tv is how I stay up to date on this subject. DeviantArt and pixiv are two other good sources. The furry community has been involved with creating 3d erotic art for decades now and is worth looking into as well even if that isn't your sort of thing yiff.party/bbs/read/21351 (you might have to do some work to access that site ]]]/fur/22069 ) and we've also had a thread about 3d here ]]2921 I'm no legal expert so I just go by what the FSF says when it comes to licensing. Blender uses a license the FSF wrote and are committed to remaining free software. The real issue nowadays is how locked down GPUs are becoming, at the rate things are going in 10-15 years buying a display that doesn't require end to end encryption with online authentication will become impossible for average people. We're already seeing the beginning of this with 'smart TVs' that have applications that can't be installed, spy on users and hijacks the screen to display its own ads.
>>4037 Thanks a lot for the links and for the commentary anon, much appreciated. >that have applications that can't be installed Am I correct in assuming you meant *uninstalled* ? That's creepy stuff about spying monitors, etc. 1984 stuff for sure. Maybe the trope of elite hackers having to use old gear to bypass blocks isn't as far-fetched as the directors made it appear? Hmm. I assume the biggest existential threat to a robowaifu market are screaming socjus harpies and their hangers-on, all trying to 'overturn the patriarchy'.
>>4038 You're right I meant there are applications that can't be uninstalled as the televisions are sold at nearly a loss and the distributors make their money by collecting data, selling ad space or bundling applications for streaming services. >I assume the biggest existential threat to a robowaifu market are screaming socjus harpies and their hangers-on, all trying to 'overturn the patriarchy'. Those people have no power whatsoever and shouldn't be of any concern. My greatest fear is how locked down computing is becoming and with the public transitioning towards a 'software as a service' model where they don't own or control anything the hardware and software for general computing that allows user ownership will become more expensive as it will be aimed towards corporate clients. We're in the beginning stages of this process when it comes to VR.
>>4039 Then do you think there is a reasonable hope of something like an 'open hardware & firmware' movement taking hold in response to this. We can always write our own control software I suppose, but creating our own chips is basically beyond any single individual's reach I'd assume. BTW, what kind of timeframe do you predict that microcontrollers and SoCs won't be available to us to use as we see fit?
>>4040 >'open hardware & firmware' movement Right now the future for such a movement is not very bright thanks to trade sanctions; https:// www.bunniestudios.com/blog/?p=5590 https:// www.techdirt.com/articles/20190731/01564742686/what-happens-when-us-government-tries-to-take-open-source-community.shtml >BTW, what kind of timeframe do you predict that microcontrollers and SoCs won't be available to us to use as we see fit? Pretty sure that's already the case with SoCs thanks to the locked down bootloaders or required proprietary graphic drivers for the Mali chips. Never been into ARM computers so I'm not certain about this. For microcontrollers if that's already happened you'll find a story about it on techdirt.
>>4041 Well, that sounds like a real blackpill tbh. But before I buy into the "It's hopeless, just give up now!" position offhand, I'll try to do more research about this topic on my own. Thanks anyway anon.
> <archive ends>
>>4021 >Windows only >No source code >700~ KB only heh looks a bit suspicious if you ask me. How does this program hold up to a GPT2 based chatbot? >>2422
>>4134 Hey there. Thanks for bringing up the concern. The source is available and I've given it the once over. Other than being old and unsupported and a bit hacky (it's C with classes, for example), I don't see anything particularly suspicious about it. It's a Windows port of the ALICE and AIML project. I'm going to leave it up for now unless something intentionally exploitative is discovered about it later on, which I don't really anticipate with it ATP.
>>4135 I suppose I should clarify this a bit better. The account owner that posted that video is a poseur, he didn't author the software, nor is it his project. Here's the most current sauce I've found, posted by Jacco Bikker (also apparently the author of the software): https://archive.org/details/WinAliceVersion2.2 The original projects was a research tool done at Carnegie-Mellon: https://en.wikipedia.org/wiki/Artificial_Linguistic_Internet_Computer_Entity Led by Richard Wallace: https://en.wikipedia.org/wiki/Richard_Wallace_(scientist) I'll assume that clarifies things Anon.
>>4136 Nice research agent. From a quick glance the program performs worse than the GPT 2 based chatbot, it only good merit is that has very good performance and doesn't take about 10-30~ seconds to respond though given its size its seems to be obvious that its vocabulary capabilities is severely limited. >The account owner that posted that video is a poseur, he didn't author the software, nor is it his project. So then all he did was modifying the lines to make it more of a typical anime character based then, I looked through its text files and oh man does it look like hell to edit it. >>4135 > I'm going to leave it up for now unless something intentionally exploitative is discovered about it later on, which I don't really anticipate with it ATP. Well that clarified then since the source code of this program is available, though I have no idea how the hell those .aiml files are used, and the readme.md the author provides is highly informative.
>>4137 Yeah, it's a throwback to the old-school 'expert systems' approach (thus probably why it was basically abandoned). The reason it performs quicker with fewer resources is that it's mostly relying on pre-architected, canned responses with very little by way of statistical processing. Which brings me to your next point: >though I have no idea how the hell those .aiml files are used That's the actual encoding mechanism for these pre-canned responses. It's an XML markup variant created by this professor to support his research project with ALICE. https://en.wikipedia.org/wiki/AIML https://github.com/drwallace/aiml-en-us-foundation-alice IMO, this entire approach is mostly a dead-end from the dark ages of AI, unless some automated way was devised to program these AIML files in advance--or some hyper-autist literally spent most of his entire life devoted to building 100'000s of response variations. Statistical approaches already are the 'future' today, and particularly once we can integrate neuromorphics along with the NLP processes.
>>4139 >It's an XML markup variant created by this professor to support his research project with ALICE. >XML Big gay. >Yeah, it's a throwback to the old-school 'expert systems' approach (thus probably why it was basically abandoned). The reason it performs quicker with fewer resources is that it's mostly relying on pre-architected, canned responses with very little by way of statistical processing. Figures it, so that's why its "intelligence" is severely limited, it's a surprise that this AI even won 3x prize for that. So if I get it right it just quickly finds a pattern of the user response and then scans over its own text files to find closet match and make a response based on that, so in short a very primitive form of chatbot. >or some hyper-autist literally spent most of his entire life devoted to building 100'000s of response variations. Sounds like a waste of time, it would be probably better to devise some kind of algorithm or uh malleable objects/entity component system that defines several aspects of how the AI should respond, or whatever those fancy terms is being used by the likes of GPT, BERT and so on. It sounds like madness to me editing thousand upon thousand of text files just to have more varied responses.
>>4140 Yep, you pretty much understand it all Anon. > it's a surprise that this AI even won 3x prize for that. It just shows you where the state of AI research in NLP was before 2005. GPGPU was just becoming an idea forming in the minds of researchers, and Nvidia hadn't released it's ground-breaking CUDA toolkits yet either. Once the iPhone opened up the smartphone market ugghh the demand for high-efficiency computation performance really began picking up steam where today TensorFlow is basically the state of the art. As is easy to tell, we still have a ways to go yet, but things are dramatically different now than the days of yore when Chomsky's ideas ruled the roost.
I think a good system would use such prepared answers as a base or as one of its systems. It would, however, create many of these responses on its own while it isn't talking. It could use other systems to think about stuff and then use prepared answers in some cases and in others to fill in blanks depending on the situation.
>>4232 I like the sound of those ideas Anon. Can you expand them with some specific details for us?
It's chinese software but the movements - the mocap and the cloth physics are so fluid: https://www.youtube.com/c/LumiN0vaDesktop/videos Seems to be in beta as its just prerendered sequences, no contextual interactivity yet.
>>4235 AIML or other chat systems store sentences and logic when to use them. Those are called by some software (runtime?). One could write software which would create responses on it's own, using other software like NLP, GPT, ... There would be more time to analyze the grammar and logic, compared to doing that only when needed. Humans also think about what they would say in certain situations ahead of time, have inner monologues, etc
>>4829 I think the idea of 'pre-rendering' responses (so to speak) might have some strong merit. Particularly if we could isolate common channels most robowaifus would go down in everyday scenarios, then there might be some efficiencies in runtime performance to be gained there.
>>4028 >best Gravity Falls episode kek
Had no idea how to 3D model when I started this but I'm slowly making progress. I just hope my topology isn't complete trash, kek. AI in Godot To get Pytorch to work in Godot 3.2.2. Start a new project, click the AssetLib tab, install the PythonScript plugin and restart the editor. In your project's folder go into ./addons/pythonscript/x11-64/bin (on Linux), make pip and python3 executable and then edit the first line of the pip script to point to the path of python3. For CPU only (Linux & Windows): ./pip install torch==1.6.0+cpu torchvision==0.7.0+cpu -f https://download.pytorch.org/whl/torch_stable.html With CUDA 9.2 support (Linux only): ./pip install torch==1.6.0+cu92 torchvision==0.7.0+cu92 -f https://download.pytorch.org/whl/torch_stable.html Now you have access to the full power of AI within Godot. The world of VR waifus is yours. To test that it's working, run Godot from the command line and add a Python script to a node like this: from godot import exposed, export from godot import * import torch @exposed class test(Node): def _ready(self): l = torch.nn.Linear(2,3) x = torch.rand(2) y = l(x) print(l(x)) It should output a result to the terminal something like tensor([-0.2603, 0.2927, -0.8231], grad_fn=<AddBackward0>) I'll be updating TalkToWaifu to make it easier to use and more modular so it can be integrated into Godot easily. Stereo imaging If you don't have a VR headset it's possible to render two cameras on screen at the same time in Godot. The left eye camera should be on the right side and right eye on the left so you can cross your eyes to look at them and see them in 3D. https://www.youtube.com/watch?v=8qGPOZW4T_M Open-source VR Also there's an open-source VR headset available too if you feel like building your own and don't wanna worry about being forced to log into the Ministry of Truth to see your waifu: https://github.com/relativty/Relativty I haven't tried it yet but it looks pretty decent.
>>5440 Neat! Thanks for the detailed instructions Anon. That's pretty encouraging to see your personal breakthrough with this. I'm sure this will be a very interesting project to track with.
>>5440 >and don't wanna worry about being forced to log into the Ministry of Truth to see your waifu You. I like you Anon.
>>5440 Ah, this looks interesting, unfortunately my weebshit making machine is windows only, I got to installing the PythonScript, which exe would I be running to get the equivalent, thanks!
>>5456 I don't have access to a Windows VM at the moment but I think all you need to do is install pip manually first. >On Windows, pip must be installed first with `ensurepip`: $ <pythonscript_dir>/windows-64/python.exe -m ensurepip $ <pythonscript_dir>/windows-64/python.exe -m pip install torch==1.6.0+cpu torchvision==0.7.0+cpu -f https://download.pytorch.org/whl/torch_stable.html https://godotengine.org/asset-library/asset/179
>>5458 That does the job, thanks! How are you going to use the AI, in human recognition? In training to create new animations? If you're going with 3D, will you still use stock animations e.g. from Mixamo, and you'll just use the AI for selecting the behavior and corresponding animation? I'm thinking of maybe trying with 2D first. I was going back to Godot to try to make a cheapass version of Live2D. When researching Vtuber tools, I noticed that the free Facerig clones were written in Unity game engine. So there is potential for Godot AI to take the place of both Live2D and Facerig: Read and process facial recognition from the webcam, then adjust the waifu puppet animation accordingly, all within the same game engine.
>>5459 Glad to hear it went smoothly. The possibilities are endless really. My first real objective is to do a walking simulation where AI controls movement of all the joints and can keep balanced even if hit with a beachball. Then I might extend it to running, jumping and picking up objects. I just wanted to create a little demo of what's possible with AI now to inspire people to get into waifudev. People could do a lot of other things with it, like facial recognition for custom Vtubers. I haven't given it too much thought yet. I'm more focused on developing AI, although I do want to create an unscripted video of my waifu giving an AI tutorial by the end of the year and I'll probably use Godot to do it. Which reminds me you can do real-time speech synthesis with models like WaveGlow within Godot: https://nv-adlr.github.io/WaveGlow
Open file (19.78 MB 1280x720 Viva Project v0.8.mp4)
>>340 Viva Project v0.8 comes out on October 31st. He just released a new video a few days ago. https://www.youtube.com/watch?v=CC4ate84BiM
Open file (176.19 KB 700x700 Relativity-VR-front.jpg)
Open file (187.12 KB 700x700 Relativty-VR-open.jpg)
Once again meta, since there also is no dedicated thread for VR yet, since only few have it and it's still quite expensive: https://www.relativty.com/ created a OpenSource VR headset which caught some attention and might lift off, while the Quest-2 requires that you share all your movement data and interactions with Facebook and also can loose all access to everything stored on that device and related to it, like bought games, contacts, save games, etc if you break any of their rules on their VR platform or on Facebook. This will most likely also include fake names or wrong infos about yourself in your profile. The one from https://www.relativty.com/ might develop as the cheaper alternative to some better ones, but it's behind in quality and probably always will be. Since it seems only to require investments around 200$ and access to a 3d printer it might be very interesting for many here. It might be possible to improve a lot and get really good. Main issue seems to be the tracking, sensors and such. It uses Pytorch and Cuda for experimental tracking. Also it needs connection to a computer, which might create annoying problems with cables, while the more expensive ones are standalone. I works with Steam, though. Github: https://github.com/relativty/Relativty
>>5782 Yes, there's some room for improvement I'm sure. But it seems like it's actually a remarkable progress so far for a hobbyist effort. 3D printing files, PCB plans, electronics & hardware lists, software, cabling, everything. Seems like an independent project by professionals. The Steam thing is a little worrying, but since they seem to be entirely open source thus far, then hopefully there won't be any entrenched botnet in the product.
>>5783 With Steam I meant they're compatible, which is clearly a plus. It doesn't mean the set is dependent on it. Wouldn't be a problem anyways, since one could just remove this part in an open source system. The problem with the Quest-2 is that the games are stored on that device and the device also is dependent an having a account. So they have full control. In a way it isn't your device, you're just paying for it. So probably you would even need FB approval of any Waifu software and maybe a (probably expensive) developer account.
>>5784 yep, no debate on those points tbh.
bump for vr
>>5788 Yeah, please don't. Build one and tell us how it went, or develop some VR waifu, then everyone will be interested.

Report/Delete/Moderation Forms

Captcha (required for reports)

no cookies?