Research map / 3D vision / Gaussian splatting
Rendering Humans behind Occlusions
Tiange Xiang, Adam Sun, Scott L. Delp, Kazuki Kozuka, Li Fei-Fei, Ehsan Adeli
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 occlusion-aware three-part scene parameterization instead of optimizing on visible parts only (ablation) | Wild2Avatar · OcMotion | qualitative rendering quality | Improved |
| add occlusion decoupling loss L_occ (ablation) | Wild2Avatar · OcMotion | qualitative recovery of occluded region | Improved |
| add geometry completeness loss L_comp (ablation) | Wild2Avatar · OcMotion | qualitative geometry completeness and pose | Improved |
| jointly optimize predicted SMPL parameters with the model (ablation) | Wild2Avatar · OcMotion | reconstruction quality vs GT SMPL | Improved |
| Wild2Avatar instead of Vid2Avatar | Vid2Avatar · OcMotion | quality_user: 3.7 -> 6.6 | Improved |
| Wild2Avatar instead of Vid2Avatar | Vid2Avatar · In-the-wild videos | comp.: 0.66 -> 0.74 | Improved |
| Wild2Avatar instead of OccNeRF | OccNeRF (500 frames) · In-the-wild videos | quality_vis: 7.98 -> 9.68 | Improved |
| Wild2Avatar instead of HSR | HSR · Synthesized occluded NeuMan videos | PSNR: 18.77 -> 22.47 | Improved |
| reduce training frames with back view of the human (ablation) | Wild2Avatar · OcMotion | qualitative geometry completeness | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Vid2Avatar — Vid2Avatar: 3D Avatar Reconstruction From Videos in the Wild via Self-Supervised Scene Decomposition (CVPR 2023) · better on 6 of 6 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
- OccFusion: Rendering Occluded Humans with Generative Diffusion Priors (NeurIPS 2024 · similarity 0.90)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.90)
- Robust Multi-Object 4D Generation for In-the-wild Videos (CVPR 2025 · similarity 0.86)
- MultiPly: Reconstruction of Multiple People from Monocular Video in the Wild (CVPR 2024 · similarity 0.86)
- IntrinsicAvatar: Physically Based Inverse Rendering of Dynamic Humans from Monocular Videos via Explicit Ray Tracing (CVPR 2024 · similarity 0.86)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · similarity 0.85)
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.
