Belgian technology company SoftKinetic is developing gesture recognition hardware and software for real-time range imaging cameras. The technology is being used for video games, consumer electronics and fitness companies. SoftKinetic's gesture recognition software platform, named iisu, can recognize and distinguish or isolate different scenic elements, can identify and track the body parts of a user, and can adapt the user's shape, posture, and movements to an existing physical model, and vice versa. Coupled with SoftKinetic's DepthSense hardware technology, it opens up new opportunities for interaction. Eric Krzeslo, Chief Marketing Officer at SoftKinetic, explains how the company is working with NVIDIA Tegra technology in this exclusive interview.
What was your vision when developing this technology?
Our goal is really to enable more intuitive and immersive interaction between a user and the computer or mobile platform, and the digital content it can run on this platform.
How do you feel your hardware innovates today’s technology?
With the 3D vision we are doing and you the gesture recognition, that really adds that new dimension to the interaction with the device. Everybody is looking for more natural and immersive interaction with these complex devices like tablets or PCs.
What do you think consumers will like most about your technology?
If the technology is invisible, then people will like it. We need to be embedded, we need to be very natural and robust. Once the technology gets to that point people won’t notice it and it will just be a new way of interacting.
How has SoftKinetic worked with NVIDIA with this technology?
We are leveraging the power of the NVIDIA GPU to run our advanced algorithms for 3D vision and gesture recognition. NVIDIA provides the CUDA API, which handles things like shaders and advanced algorithms.
How does your technology make use of the latest Tegra hardware?
It’s really the raw power of the GPU that really helps us to run very advanced gesture recognition and 3D vision algorithms that would run on the mobile platform otherwise.
What are the challenges of developing for Android devices today?
Android devices are very open, which is a good thing. But it’s so open that it’s difficult to find your way through the content and different use cases that you can have on Android devices.
How does NVIDIA help with this?
NVIDIA is trying to narrow down the ecosystem, or is adding a layer on top of Android to have a better focus on the experience you can have on their devices.
What are your thoughts on NVIDIA Tegra K1 technology?
Tegra K1 is really an amazing GPU in terms of raw processing performance that we really need for what we are doing with our 3D camera.
What do you feel Tegra opens up for you as a hardware and software developer?
It really brings the mobile platforms to the level of the most high-end work stations like PC or all game consoles. That’s really interesting for us because we all know that the future is in mobile, so it’s really great to have that processing power available on the mobile platform.
What type of leap is there between what you can do with Tegra 4 and what’s possible with Tegra K1?
Already with the Tegra 4 we had this amazing performance, but with the K1 so far what we’ve seen is that it’s really an order of magnitude higher than the Tegra 4.
What are your thoughts on the speed at which NVIDIA mobile technology is accelerating?
That’s quite amazing. I’m not sure about the time between the T4 and the K1, but I think it’s something like eight months. That’s a very short cycle and we can’t wait to see the next iteration of their GPU.