Uthana’s cover photo
Uthana

Uthana

Technology, Information and Internet

Los Angeles, CA 2,152 followers

Uthana builds foundation models for human motion. Any character, controllable in real-time.

About us

Uthana builds foundation modes for human motion. Any character, controllable in real-time. We support animators, game developers, avatar creators, artists, riggers, and 3D generalists. Created motion faster than ever.

Website
https://www.uthana.com/?utm_source=linkedin
Industry
Technology, Information and Internet
Company size
2-10 employees
Headquarters
Los Angeles, CA
Type
Privately Held
Founded
2023

Locations

Employees at Uthana

Updates

  • Uthana reposted this

    I'm excited to announce that Uthana’s AI-powered animation is now live in Unity AI. Unity recently launched Unity AI in Open Beta, with contextual assistance, asset generation, and curated third-party models directly inside the Editor. Starting today, Uthana brings the motion layer to that stack. Unity developers can now access Uthana’s text-to-motion and video-to-motion inside Unity AI: -- Text-to-motion turns a prompt into production-ready character animation in as little as 3 seconds -- Video-to-motion captures the performance of a reference video with high fidelity, including finger detail and foot contact -- Both work across any bipedal character As AI-assisted coding and agentic workflows improve, creative motion is still one of the hardest bottlenecks in building games. Uthana helps close that gap inside one of the world’s leading game creation platforms, helping creators move faster while retaining control of the creative process. Uthana is an AI platform for human motion. We build proprietary models that understand how the human body moves in 3D. Our tech is grounded in high-quality motion capture data and the underlying rotations, positions, and joints that make movement feel real. That foundation lets us recreate lifelike motion with high fidelity in a fraction of the time traditional workflows require. We’re starting with text-to-motion and video-to-motion, with additional animation and rigging capabilities planned over time. Stay tuned! #AIforGames #3Danimation #GenAI

  • Shaun Seong-young Kim demonstrates what can be done with video-to-motion 2.0. What are you creating?

    1. Unlike standalone Gen AI video, in-game cutscenes must blend seamlessly with gameplay cameras and utilize native level assets, requiring a fundamentally different approach. 2. Fake Scenario: A cinematic "Quest Giver Encounter" to test a new rapid In-Game Previs pipeline. 3. Caveats (Speed > Polish): Voices are temp tracks generated via ElevenLabs. AI Mocap data is completely raw to showcase out-of-the-box quality, with no manual polishing or Control Rig adjustments. 4. Direct to Shot Design: Traditional previs is bottlenecked by heavy asset prep (rigs, environments, voice files) and spends over 50% of the time on character blocking. This workflow bypasses those hurdles, allowing an immediate focus on staging, camera work, and visual storytelling. (Total time: 8 hours (2 hours/day for 4 days), covering shot design, basic lighting, retargeting, a collapsing sim, and motion/voice/SFX generation.) 5. V2V Mocap & Text-to-Motion (feat. Uthana): Thanks to Uthana CEO Viren Tellis (met at GDC) for the test account. Beyond V2V capture, their Text-to-Motion instantly generated missing micro-movements (e.g., crouching to stand). A future web-based curve editor to clean-up the data would make this even stronger. 6. Conclusion: With a dedicated TA handling Unreal Engine retargeting and temp asset generation (via GPT, Gemini 3 Flash Image, Tripo 3D, Accurig), this pipeline could be 5-10x faster than traditional methods. It is highly viable for immediate previs use. The necessary next step is establishing cross-departmental guidelines to ensure these temp setups transition smoothly into final production assets. #previs #ingame #AI #mocap

Similar pages

Browse jobs

Funding

Uthana 2 total rounds

Last Round

Seed

US$ 4.3M

See more info on crunchbase