Kinetix

Kinetix is an AI research lab for 3D motion capture, text-to-animation, retargeting, rigging, and physically grounded AI video generation.

AI 3D 모델 생성기 AI 동영상 생성기 텍스트-동영상 변환

웹사이트 방문

Research lab for 3D human motion and grounded video generation

Kinetix is an AI research lab focused on 3D human motion intelligence. Its work combines machine learning and 3D engineering to understand, generate, and control humanoid motion for character animation and related media workflows.

The site highlights two main research directions: 3D animation technologies and AI video with physical grounding. On the 3D side, Kinetix develops markerless motion capture, text-to-animation, mesh-aware retargeting, and automated rigging. On the video side, it presents Kamo-1, a 3D-conditioned video generation model that uses spatial and motion priors to steer camera and character movement together.

Core capabilities

Markerless full-body motion capture

Reconstructs temporally consistent, lifelike movement from raw acting footage using a proprietary model and body model, producing 3D animation data from a single camera input.

Text-to-animation generation

Generates full-body 3D animation from natural language prompts while keeping motion coherent over time.

Mesh-aware retargeting

Adapts source motion to a target character’s morphology by modeling skeletal structure and mesh geometry, helping reduce clipping, interpenetration, and other mismatches.

Automated rigging

Automatically infers an adapted skeleton for unrigged characters using geometric and anatomical priors, creating a rig suitable for downstream animation.

Physically grounded AI video generation

Conditions video generation on 3D priors such as camera trajectories, reconstructed 3D character animations, reference images, and natural language prompts to control both camera and character motion.

Data and training pipeline

Works from a large proprietary motion dataset and a synthetic scaling pipeline built to expand training data across many character morphologies, scenes, lighting conditions, and camera settings.

Practical workflows and scenarios

Capture motion from live footage
Use Kinetix’s motion capture research when converting raw acting footage into temporally consistent 3D motion data for animation pipelines.
Generate and prepare character animation
Use the text-to-animation and rigging technologies when generating or preparing character motion for 3D production work.
Transfer motion across characters
Use mesh-aware retargeting when motion needs to move between characters with different body shapes or skeletal structures.
Create physically grounded AI video
Use Kamo-1 when camera movement and character movement need to be coordinated inside the same generated video shot.
Embed motion capabilities into products
Use the team’s technologies as embedded capabilities in tools or game workflows, as shown by integrations with Unity Muse, OVERDARE, and Adobe Mixamo.

Pros and Cons

Pros

Covers the character animation pipeline from rigging and retargeting through motion generation.
Supports multiple motion inputs and control signals, including single-camera footage, text prompts, images, and 3D camera trajectories.
Uses geometric and anatomical priors to handle character-specific constraints more carefully than a generic animation model.
Backed by published research and examples of real-world use in Unity Muse, OVERDARE, and Adobe Mixamo.
Includes a physically grounded approach to video generation rather than treating motion as unconstrained pixels.

Cons

The public pricing page currently returns a 404, so there is no usable pricing information in the collected sources.
The site does not provide a general integration list, API reference, or detailed product packaging in the collected pages.
Several claims are framed at a research level, so readers may need additional documentation to understand availability and deployment details.

FAQ

What does Kinetix do?

Kinetix is an AI research company focused on understanding and generating human motion in 3D. Its work spans motion capture, text-to-animation, mesh-aware retargeting, and AI video with physical grounding.

Is Kinetix a consumer app or a research lab?

The source material presents Kinetix primarily as a research lab and technology provider rather than a packaged SaaS product. It builds models and tools for 3D animation and human motion workflows.

What kinds of motion technologies does Kinetix work on?

Kinetix’s 3D research page highlights markerless full-body motion capture from a single camera, natural-language-to-full-body 3D animation, mesh-aware retargeting, and automated rigging.

Does Kinetix list pricing?

The pricing page currently returns a 404, so the site does not provide a public pricing model in the collected sources.

Which products or platforms has Kinetix worked with?

The site references integrations and deployments with Unity Muse, OVERDARE by KRAFTON, and Adobe Mixamo, but no general integration list or API documentation is shown in the collected pages.

Quick Facts

Category: AI research lab / 3D animation
Primary focus: Human motion intelligence, character animation, and AI video
Notable model: Kamo-1
Website: kinetix.tech
Pricing page: 404 / not available in collected sources
Example workflows: Motion capture, text-to-animation, retargeting, rigging, grounded video generation