3 ms·

I must have missed the perfect hand tracking. Could you share where it was? All I saw was rather exaggerated movements to make things work.

felixvolny 12y ago
This requires perfect hand tracking: https://i.imgur.com/QzSD0bq.png https://i.imgur.com/QzSD0bq.png
- teraflop 12y ago
  I think the gun is a real, physical object -- I mean, it even casts shadows on the user's arm. It's probably just the enemies, HUD and glowy, flashy special effects that are rendered in VR, and those should be more forgiving of slight positional inaccuracies.
  - duffmancd 12y ago
    Especially as it comes from Weta workshop who originally designed the weapon (https://wiki.teamfortress.com/wiki/Righteous_Bison https://wiki.teamfortress.com/wiki/Righteous_Bison) - if anyone has a physical copy it's them
    - mwilcox 12y ago
      WETA have been selling physical replicas of these weapons for years: https://www.wetanz.com/rayguns/ https://www.wetanz.com/rayguns/
m_eiman 12y ago
Grabbing and handling the gun.
- nknighthb 12y ago
  The gun is a physical object.
teraflop 12y ago
The interface widgets appear to be occluded by the user's hand, with no visible artifacts even when the hand is moving quickly. That means the system has to be tracking the hand and somehow separating it from the background, with latency of much less than a single frame. EDIT: here's a particularly blatant example, around 0:17: http://i.imgur.com/8mSv8oC.jpg http://i.imgur.com/8mSv8oC.jpg The user is swiping from right to left, causing their hand to be highly motion-blurred. But the rendered UI elements smoothly fade in behind the hand. There are foreground segmentation algorithms that can do this kind of thing, assuming you have some information about which regions of the image are foreground/background (say, from a depth sensor). But those algorithms typically take on the order of seconds or minutes to process a single frame. For this video to be real, the total latency -- sensing, matte generation, and rendering -- would have to be in the single-digit milliseconds. It's barely on the edge of plausibility, but I'm very very skeptical.
- nknighthb 12y ago
  OK, that's right. I'm still not convinced anyone can say that this must be faked. None of this is impossible with current tech, just not likely in a wearable/battery-powered device. Edit: Actually, no, I don't think it's right. Modern chroma keying could probably handle this.
- MichaelGG 12y ago
  If you had head mounted cameras, they could feed depth info to layer the UI, as you say. And if the display is covering your eyes completely, then feeding video in...? We don't get to see what the guy looks like, just what he's seeing. Maybe he's got 15lbs of Heath Robinson equipment on his head.

felixvolny 12y ago

This requires perfect hand tracking: https://i.imgur.com/QzSD0bq.png https://i.imgur.com/QzSD0bq.png

teraflop 12y ago
I think the gun is a real, physical object -- I mean, it even casts shadows on the user's arm. It's probably just the enemies, HUD and glowy, flashy special effects that are rendered in VR, and those should be more forgiving of slight positional inaccuracies.
- duffmancd 12y ago
  Especially as it comes from Weta workshop who originally designed the weapon (https://wiki.teamfortress.com/wiki/Righteous_Bison https://wiki.teamfortress.com/wiki/Righteous_Bison) - if anyone has a physical copy it's them
  - mwilcox 12y ago
    WETA have been selling physical replicas of these weapons for years: https://www.wetanz.com/rayguns/ https://www.wetanz.com/rayguns/

m_eiman 12y ago

Grabbing and handling the gun.

nknighthb 12y ago
The gun is a physical object.

teraflop 12y ago

The interface widgets appear to be occluded by the user's hand, with no visible artifacts even when the hand is moving quickly. That means the system has to be tracking the hand and somehow separating it from the background, with latency of much less than a single frame. EDIT: here's a particularly blatant example, around 0:17: http://i.imgur.com/8mSv8oC.jpg http://i.imgur.com/8mSv8oC.jpg The user is swiping from right to left, causing their hand to be highly motion-blurred. But the rendered UI elements smoothly fade in behind the hand. There are foreground segmentation algorithms that can do this kind of thing, assuming you have some information about which regions of the image are foreground/background (say, from a depth sensor). But those algorithms typically take on the order of seconds or minutes to process a single frame. For this video to be real, the total latency -- sensing, matte generation, and rendering -- would have to be in the single-digit milliseconds. It's barely on the edge of plausibility, but I'm very very skeptical.

nknighthb 12y ago
OK, that's right. I'm still not convinced anyone can say that this must be faked. None of this is impossible with current tech, just not likely in a wearable/battery-powered device. Edit: Actually, no, I don't think it's right. Modern chroma keying could probably handle this.
MichaelGG 12y ago
If you had head mounted cameras, they could feed depth info to layer the UI, as you say. And if the display is covering your eyes completely, then feeding video in...? We don't get to see what the guy looks like, just what he's seeing. Maybe he's got 15lbs of Heath Robinson equipment on his head.