Gracia AI’s Post

View organization page for Gracia AI, graphic

1,072 followers

What we do and why we do it VR lacks life-like experiences. That's what we address at Gracia delivering feeling of presence through photorealism and natural parallax effects in our 3D scenes. Look at the level of details! How it works: Step 1. The scene or an object is captured using regular 2D cameras from a few viewpoints. Any lighting conditions, and challenging surfaces. Step 2. We train an AI model from scratch based on the footage on our proprietary pipeline. As a result we get an AI model that "knows" how the scene looks like from any viewpoint. Step 3. On-device rendering. One can walk inside the scene and experience it from any angle and perspective while AI model renders the picture for both of your eyes separately in the real-time (550 fps actually). Separate picture for both eyes delivers this parallax effect, which our brain uses to understand the distance and size of an object. To the best of our knowledge, we are currently the only project that enables real-time rendering on standalone headset ;)

To view or add a comment, sign in

Explore topics