IVN gets rid of cartoony avatars

Companies interested in the cost-savings and productivity benefits of immersive virtual environments but put off by the cartoony avatars may soon have another option, according to LA-based Integrated Virtual Networks.

IVN’s patented “Silhouette” technology uses standard cameras to capture the user’s image, isolates it from the background, and converts it into a live, 3D avatar.

Company spokesman Craig McAllister told Hypergrid Business that the company has been in quiet development mode for the past three years, but is now ready to go public, and is now looking for investors and business partners.

Rollout plans

The platform will go live before the end of the year, McAllister said. “But it could be sooner.”

The Silhouette avatars are currently only available for use with the company’s own Nexos virtual environment — a pre-built, 3D business park.

Nexos. (Image courtesy IVN.)

The company is open to licensing the Silhoutte technology to other virtual environment vendors — enterprise-focused firms like Protosphere, or gaming or virtual education companies. Integration with open source environments like OpenSim and Open Wonderland is possible as well.

“It’s all doable,” he said. “We can pretty much do whatever the market dictates.”

Pricing is still to be determined.

Image quality

In the demo video, movements, gestures and facial expressions are rendered quickly — but there is sometimes a white edge around the avatars.

This will be cleaned up before the final release, McAllister said.

“But for the video, we wanted to make sure that the line was visible,” he said. “We wanted it to be very clear that this is being done in-world. This is not a produced thing.”

He added that the image capture technology works with standard technology — a high speed Web camera and broadband connection, and a video card with 3D graphics acceleration.

McAllister said that he’s sat in on meetings with half-dozen simultaneous attendees, but couldn’t say what the maximum capacity of the platform was, though this information will be available later as well.

Two types of avatars are available on the platform — a full-body avatar, which can be either sitting or standing, and a head-only avatar which can be combined with a computer-generated body.

Unlike traditional video conferencing, this technology creates the feeling of sharing a common space with the other attendees, and also creates the illusion of eye contact.

And, unlike cartoon-style avatars, facial expressions and gestures are accurately transmitted in real time, which can be important not only in business meetings but other contexts as well, such as virtual therapy and counseling sessions, poker games, and virtual education.

Related Posts


Maria Korolov

Maria Korolov is editor and publisher of Hypergrid Business. She has been a journalist for more than twenty years and has worked for the Chicago Tribune, Reuters, and Computerworld and has reported from over a dozen countries, including Russia and China. Follow me on Twitter @MariaKorolov.

6 Responses

  1. For most applications, I like the idea of using an actual face on an avatar, but would not favor using the whole body.  As the video illustrates, the full body avatar requires the person to stand and move in an actual space to give the avatar motion, which would add significant complexity for participants, as actors who have worked on green screen sets have often noted in interviews.

    Nonetheless, I see whole body and even face only video insertion as an intermediary step.  If facial recognition software becomes sufficiently advanced, an avatar’s face could be synced to a real face.  This allows for an emphasis on communication over presentation.  Whether a person is photogenic, or having a “bad hair day” should not detract from interactions, and is a dimension worth stripping out of most distance communications.  That was always the good thing about phone conferencing, even at the cost of missing facial expressions.  What avatars lack is the sophistication of showing accurate eye movements and expressions, such as agreement or confusion.  Once they have those qualities, I think the avatar will gain much broader acceptance, even if they are not entirely photorealistic.

  2. joey1058@gmail.com' Joe Nickence says:

    IVN has been promising some version of this tech for the last several years without actually releasing any of it.  The “last three years” stuff is smoke and mirrors.  I’m not going to get real excited over this announcement, considering how they’ve handled their other properties.  I actually want to see people using this, and not keep watching updated videos of what it can do.

  3. guest@guest.com' Guest says:

    ivn…. arent you dead?

  4. Ener Hax says:

    i dunno – it’s like Unity and Jibe – everyone thinks it’s more realsitic and it is but it is still a far cry from Hollywood level or big game graphics – i’d rather stay cartoony and easy to use over becoming a multimillion dollar enterprise

  5. Han Held says:

    Uncanny Valley, anyone?

  6. d3vicha@gmail.com' Donavan says:

    Will this kind of avatar be able to change clothes or use accessories built in the virtual world? What happens to that infrastructure that economically supports it? The whole business model is going to have to adapt to that aspect of avatar generation. Not sure that’s viable when you add the expense of how the avatar is generated.

    The presence of cartoony avatars is already there. Making them more real doesn’t add to that. I’ll agree in a face to face situation, it would be nice to play “Lie to Me” facial games but is that really worth the expense?