Philip Rosedale: FOVE's Eye-Tracking Tech Addresses the "Most Important Missing Component" for VR Communication

FOVE: The World's First Eye Tracking Virtual Reality Headset, is a new Kickstarter VR project which incorporates eye-tracking technology into a VR head-mounted display (or HMD) to very cool effect -- watch:

Co-founded by Lochlainn Wilson and Yuka Kojima, the project has a very prominent supporter: Philip Rosedale, who tells me he plans to make his new VR-compatible world High Fidelity compatible with the FOVE. In fact, he believes the eye-tracking technology is so important for the future of VR, it'll become standard in the next generation of VR HMDs:

"Being able to see exactly where someone else is looking is probably the most important missing component of 1:1 communication in a virtual environment, if we have head and upper body motion, facial gestures, and hand movement," as Philip puts it to me. "Eye contact is extremely important and eye movement is a rich form of communication. Having used it, I can attest to the fact that FOVE has demonstrated successful, accurate eye tracking in an HMD. So we can expect that second generation HMD's will have eye tracking built in, and that this will be a big and important advance in 1:1 presence."

That sounds right to me. With Philip's first VR world, Second Life, the avatar eye-lock function is an incredibly compelling way to create simulated connectivity between two people. Go here to consider getting in on this Kickstarter.

Thanks to Adrian Cutler for the tip!

Please share this post:

Tweet

 

Comments

[Sepp]

But, but, but, "depth of field" is someting that apply to photography. You don't "see" depth of field in real life, not in the same sense at least, the brain takes care of that. Using eye tracking technology to add depth of field in realtime will not make for a more realistic experience, on the contrary it will be very artificial indeed.

[Meursault]

A recent glassdoor review for Linden Lab left this gem: "Oh, well. Thanks for paying me three months to spend time learning Sphinx speech recognition, 3D audio and phoneme"

So it looks like LL is working on voice to viseme (mouth shape) translation. It's not that hard - there are only a dozen mouth shapes in english and a company like iSpeech can give you a timeline of viseme patterns to any audio.

Not sure about eye gaze. How exactly is it the #1 missing communication? NWN tends to blow the trumpet a lot. IMO, since the face is such a high fidelity communication device, High Fidelity's work on mapping facial expressions to avatar rigging is more compelling as it expands the vocabulary of virtual expression. I bet HF has hit a wall where human-like facial expressions look super creepy if the eyes are not part of the overall expression.

The original eye-gaze research is at least 15 years old. Has anyone demonstrated that it creates a more immersive experience? Think of a Skype call or Hangout - is eye contact really important there?

The one thing that eye gaze tracking can do, is give advertisers super high fidelity data on their advertisements. One SL viewer already did eyeball tracking (POV/view tracking) back in 2006.

[Kitty Revolver]

I love Rosedale, but $400 bucks is a lot to put down on a headset, plus the Hydra gloves, you are spending upwards of $1000 to really do VR. That just won't fly in the mass market and will likely kill High Fidelity quickly if Rosedale places all his bets into these technologies to make High Fidelity really "sing." Personally HiFi promises a lot and doesn't seem to properly deliver (yes yes alpha, but still). I think his focus is too techie for mass market.

[Kim Anubis]

Even if It isn't someone's eyes I usually watch while conversing, I find all of this work on facial tracking and avatar rigging very inspiring, because someday maybe I will be able to lip read inworld.