GUSH3R: Photorealistic Dynamic Human Scene Reconstruction from Monocular Video
janusch_patas · x · 2026-07-07
The paper introduces GUSH3R, tackling the novel problem of feed-forward, photorealistic, and renderable dynamic human scene reconstruction from monocular video. Core contributions include an architecture bridging human scene foundation models with photorealistic rendering, utilizing geometric priors and SMPL-X representations. It achieves competitive novel view synthesis quality compared to decomposition-based and optimization-based baselines while significantly boosting inference efficiency.
Related event: GUSH3R Enables Dynamic 3D Reconstruction from Monocular Video(2 posts)→
More from Multimodal
- FastH3-Live hits 22fps: acceleration node benchmarks and the --vram-headroom trick — spartong945 · 2026-09-11
- Midjourney style code share: --sref 2912175708 — tisch_eins · 2026-09-11
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11
- MiniMax H3 MAX nails cooking anime clips: 15-second curry demo with prompts shared — Hailuo_AI · 2026-09-11
- MiniMax Music Production Toolkit 2.5 for ComfyUI adds full mastering chain — Vivid_Promise1700 · 2026-09-11
- New Node Finder for ComfyUI ranks fresh nodes by star velocity and recency — Luke2642 · 2026-09-11