Research map / 3D vision / Face and shape
AHOY! Animatable Humans under Occlusion from YouTube Videos with Gaussian Splatting and Video Diffusion Priors
Aymen Mir, Riza Alp Guler, Xiangjun Tang, Peter Wonka, Gerard Pons-Moll
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add hallucination supervision and full pose-dependent avatar stage over coarse avatar (ablation) | AHOY · BEHAVE novel view | PSNR: 16.10 -> 24.12 | Improved |
| add hallucination supervision and full avatar stage over coarse avatar (ablation) | AHOY · BEHAVE novel pose | PSNR: 14.30 -> 22.81 | Improved |
| add RF-Inversion hallucinated video supervision (ablation) | AHOY · BEHAVE novel view | PSNR: 21.10 -> 24.12 | Improved |
| add RF-Inversion hallucinated video supervision (ablation) | AHOY · BEHAVE novel pose | PSNR: 19.40 -> 22.81 | Improved |
| add RF-Inversion hallucinated video supervision (ablation) | AHOY · BEHAVE novel view | LPIPS: 0.143 -> 0.090 | Improved |
| add map-pose/LBS-pose decoupling (ablation) | AHOY · BEHAVE novel view | PSNR: 22.60 -> 24.12 | Improved |
| add map-pose/LBS-pose decoupling (ablation) | AHOY · BEHAVE novel pose | PSNR: 20.60 -> 22.81 | Improved |
| add map-pose/LBS-pose decoupling (ablation) | AHOY · BEHAVE novel pose | LPIPS: 0.152 -> 0.110 | Improved |
| add head/body split supervision (ablation) | AHOY · BEHAVE novel view | PSNR: 23.60 -> 24.12 | Improved |
| add head/body split supervision (ablation) | AHOY · BEHAVE novel pose | PSNR: 22.20 -> 22.81 | Improved |
| give baselines canonical-pose input instead of occluded input (LHM) | AHOY · YouTube animation | PSNR: 19.03 -> 20.17 | Improved |
| give AHOY canonical-pose instead of occluded input | AHOY · YouTube animation | PSNR: 22.81 -> 22.83 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- IDOL — IDOL: Inertial Deep Orientation-Estimation and Localization (AAAI 2021) · better on 6 of 6 shared results
- SyncHuman — SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction (NeurIPS 2025) · better on 4 of 4 shared results
- PSHuman — PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior (CVPR 2025 · similarity 0.90)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.89)
- GAF: Gaussian Avatar Reconstruction from Monocular Videos via Multi-view Diffusion (CVPR 2025 · similarity 0.89)
- AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction (CVPR 2025 · similarity 0.89)
- Real-time High-fidelity Gaussian Human Avatars with Position-based Interpolation of Spatially Distributed MLPs (CVPR 2025 · similarity 0.88)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
