Why The VR Shows only comprised of new girls only. new girls vr is ok, but not as exiting to buy as if it was VR's from our favorite Past Models. i wouldnt think twice buying those shows.
I could think of a list, but it would be too long.
Maybe A Few
Melisa, Alyssia Kent, Paulina, Ava, Niemira, Serena Wood, Stacy Cruz, Sybil, Tania R.
Viola, Hilary C, Lesya, Leona Mia, Ms Lynna, Tiffany Tatum
Merry Pie, Nancy A, Nasita
Demi Fray
Katya Clover
Marina Visconti
Lucy Li
Elsa Jean
Gloria

I think that VR consists of a double image, one for each eye. They didn't film the past models with that sort of camera. Software exists to run 2D film into a VR headset, but it's still essentially 2D. Software does exist to simulate 3D by shuffling the frames to produce a double image that looks 3D, but it relies on shunting background objects around and the iStripper cards don't have background.

Maybe somebody can train an AI to work with the naked female body to produce 3D. Now there's a project!

@pantalone, go look in the fullscreen scenes thread, i created several psudo 3d shaders to be viewed on 3dtv's and vr headsets (or you can playback on a normal tv and cross your eyes). I get the depth information from the brightness (front of the body is more brightly lit than the sides) from that i offset the pixels according to their brightness(the brighter they are the more they are shifted), shift left for one eye and shift right for the other. Clothing does mess with the effect though.

Not sure that's what the op was asking for though. More a newly filmed new show from those girls.

http://www.istripper.com/forum/thread/37912/1?post=566424

Nie masz uprawnień do przeglądania tego tematu

@Z22

I'd forgotten about the shader thing! I do follow the Fullscreen thread, but only vaguely because I don't understand most of it. I tried the boss-eyed thing when you posted the images a while back, but couldn't make it work. Maybe it's my brain or maybe it's just my varifocals.

As, you say, it a shader seems to produce pretty good pseudo-3D for skintones, but the clothing gets in the way. Which it often does when I look at pictures of young women. My first reaction to the OP was "no way", but then the AI thing occurred to me. It actually shouldn't be that hard to develop a working rule-base, given all the work that's gone into image recognition for self-driving vehicles. And accuracy isn't needed for digital films, as the viewer's brain will do a lot of the work. iStripper clips would be an ideal place to start to train an AI to convert legacy digital stock, because they are background-less and the subjects, as my wife points out, are pretty much all the same.

Usually, by the time things like that occur to me, somebody's already done it. So maybe we can look forward to all our favorite old movies coming soon to our VR headsets in glorious pseudo-3D! At last, a possible ***** app for VR...

Yeh, there are several methods already but most are for creating a 3d model or capturing motion for transfer to another model. There's no reason why it couldn't be done. eg:- https://www.youtube.com/watch?v=FZZ9rpmVCqE is creating a depthmap from a 2d image, and creating missing information, like the sides of the girl.

I did think of a way a while back by combining it with my recolouring shader as it seperates the clothes from the body so you can do things like making just the skin green. hmm.... just thought of another way that might work...

To do a propper job of it I guess that unless you have a insane gpu you couldn't do it realtime though so totem would have to do it themselves, not sure there is much in it for them or at least not enough to make it worth the time and equipment, like using ai to upscale the older videos.

@Z22

The vid that you posted highlighted some issues that are important if you've got a background, but your shader examples are from clips without a background. I think that a standard depthmap routine won't hack it - it would return just two shades, girl and no-girl. The depthmaps in the vid are treating the figures as flat planes and the background gives the 3D effect.

My point about the AI was that you might be able to teach it to map a simple wireframe onto the each frame of model, then move the pov on the wireframe and fit the pic of the model back on. You'd need a routine to fill in the gaps, but that shouldn't be too hard. Then you'd have a frame with two images of the model from slightly different povs.

It's not a GPU job, clearly, and you'd need to work with the raw mp4 files. But Totem currently have an almost worthless back catalogue of medium res vids and such a process would give it new life, commercially speaking. Just musing....

@pantalone

If you google "3D video converter" you will find plenty of programmes that can create 3D videos from 2D ones. I have tried a couple of the free ones on porn films and they worked quite well despite any background that was present. I have not continued to use them because I am not very interrested in actually watching 3D movies or in VR, though I was interrested in the concepts behind it.

@TheEmu

I did as you said and you're right. There are dozens of 3D convertors. I, of course, have no way of testing whether or not they're any good, because I refuse to invest in a VR Hat until there's some content that meets my stringent criteria. Angelika Greys pole-dancing naked would do it.

Totem already offers videos for most of the performers in the op's list and they have certainly made videos for other porn sites. So getting to 3D is already possible - you just have to find which convertor you like most.

But I still like the idea of an AI clonking through old card clips, reproducing every curve and nuance, every eyelid flutter and every come-hither smile in glorious hi-res 3D. Maybe one day...

@pantalone - you can test them using nothing more than a cheap pair of red-green or red-blue glasses at a cost of only a few cents. (or you can make yourself a pair from a couple of suitably coloured sweet wrappers and a bit of cardboard. You would probably want something better for seriously watching any converted video, but red-green or red-blue galsses are quite adequate to test the programs.

No AI is needed - just simple fixed function algorithms.