Kinetix

What is Kinetix?

Kinetix is a research lab at the forefront of combining 3D technology and machine learning to understand and generate human motion. The company specializes in building state-of-the-art generative models and core technologies for character animation, focusing on human pose estimation, computer vision, and foundational 3D character workflows. Their flagship product, Kamo-1, is a 3D-conditioned video generation model designed to bring physical grounding and precise control to AI video generation. Kamo-1 supports multimodal conditioning, including 3D camera trajectories, 3D character animations reconstructed from RGB acting videos, reference images, and natural language prompts. By jointly conditioning on both camera trajectories and 3D animations, it enables coordinated control over camera motion and character motion within the same generated shot. Unlike traditional video generation that treats scenes as unconstrained pixels, Kamo-1 uses depth and pose estimates as control signals, producing videos consistent with underlying spatial layouts and articulated motion. The model can reconstruct scenes, place 3D character meshes, animate them over time, and render final videos via a flow-matching generation process. Kinetix’s technology has been integrated into Unity’s Muse AI services, allowing game developers to generate 3D animations from videos. Additionally, Kinetix powers experiences like Emote Creator for OV, enabling users to create expressive character animations. The lab's research extends to physical grounding in AI video, ensuring generated motions adhere to real-world physics. Use cases include game development, virtual production, filmmaking, and interactive applications where realistic character animation and camera control are essential. Technical details involve advanced machine learning models that process 3D data to generate coherent video sequences. Kinetix aims to revolutionize how digital characters are animated, making high-quality motion generation accessible to creators.

Use Cases

Key Features

3D conditioned video generation human motion understanding multimodal conditioning support camera trajectory control character animation reconstruction pose estimation integration spatial layout consistency articulated motion rendering

Who is it for?

AI researchers, 3D animators, computer vision engineers, game developers, film production studios, VFX artists, robotics teams

Similar Tools