Unverified50% confidenceBenchmarkExact time
HORIA achieves state-of-the-art performance on the task of 4D reconstruction of human-object interaction from monocular video.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/20/2026
First Seen
Valid until: 11/3/2026
Sources
Related Entities
Related Claims
UnverifiedMonocular 4D reconstruction refers to simultaneously recovering the shapes of the human and the object in 3D space and their motion over time from video captured with only a single ordinary camera.66% similarUnverifiedHORIA's interaction-aware re-rendering pipeline consists of three steps: rendering the preliminary 4D reconstruction of the human and object from novel views, using a text-to-video diffusion prior to refine the rendered results conditioned on the stage text, and feeding the refined renderings back as auxiliary supervision for the 4D reconstruction.64% similarUnverifiedThe standard workflow involves four-step generation and is suitable for most users; the fourth workflow is for image-to-video or first-and-last-frame video; the fifth is for reference image video generation to maintain character identity consistency.64% similarUnverifiedGen4P proposes that a pretrained video generator can be repurposed into a feed-forward 4D perceiver that recovers 4D geometry and motion information from monocular video.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073811API
curl https://kongchang.com/api/v1/knowledge/claims/1073811MCP
get_claim(id=1073811)