Unverified60% confidenceFactExact time
Atlas can accept multi-modal inputs including text, images, video, and 3D data, and uniformly convert them into controllable high-definition video
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
9/5/2026
First Seen
Valid until: 12/4/2026
Sources
Related Entities
Related Claims
UnverifiedIn sprite sheet animation, individual frames are typically packed into a Sprite Atlas/Texture Atlas to reduce GPU texture-switching overhead and improve rendering performance.70% similarUnverifiedAtlas adopts a generative Real-to-Sim approach, actively generating the RGB images and depth data that the robot's onboard camera should observe along a specific motion trajectory as the simulated robot moves through the scene.69% similarUnverifiedThe visual model receives each screenshot and outputs structured information, including OCR-recognized text, page layout, pixel coordinate bounding boxes of UI controls, and element states.63% similarUnverifiedThe visual reverse-inference module can reverse-engineer reference images or videos into positive prompts, supporting JPG/PNG images and MP4/MOV/MKV videos (up to 500 MB), with the entire process handled by a local model so that images never leave the user's machine.63% similarUnverifiedV4-Flash-Vision-Exp can read screenshots, parse charts, and complete multimodal Agent tasks in combination with tool calling.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/857952API
curl https://kongchang.com/api/v1/knowledge/claims/857952MCP
get_claim(id=857952)