Research map / 3D vision / 3D generation
RHINO: Reconstructing Human Interactions with Novel Objects from Monocular Videos
Lixin Xue, Chengwei Zheng, Georgios Paschalidis, Chen Guo, Manuel Kaufmann, Juan Zarate, Dimitrios Tzionas
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 camera-object motion disentanglement (ablation) | RHINO · Motion disentanglement ablation | Chamfer distance (cm): 10.21 -> 2.65 | Improved |
| add camera-object motion disentanglement (ablation) | RHINO · Motion disentanglement ablation | F1 @2cm (%): 26.32 -> 56.16 | Improved |
| add camera-object motion disentanglement (ablation) | RHINO · Motion disentanglement ablation | PSNR: 22.89 -> 25.80 | Improved |
| add contact-based pose refinement (ablation) | RHINO · Contact refinement ablation | penetration depth (cm): 1.088 -> 0.477 | Improved |
| add contact-based pose refinement (ablation) | RHINO · Contact refinement ablation | contact recall (%): 18.39 -> 63.57 | Improved |
| add contact-based pose refinement (ablation) | RHINO · Contact refinement ablation | contact precision (%): 24.06 -> 25.67 | Improved |
| use MASt3R-feature object pose instead of SuperPoint+SuperGlue (ablation) | RHINO · Object pose estimation ablation | F1 @2cm (%): 60.06 -> 91.38 | Improved |
| use MASt3R-feature object pose instead of LoFTR (ablation) | RHINO · Object pose estimation ablation | Chamfer distance (cm): 3.97 -> 1.09 | Improved |
| use RHINO instead of InterTrack | RHINO · Shape reconstruction (object) | Chamfer distance (cm): 11.16 -> 1.21 | Improved |
| use RHINO instead of HOLD | RHINO · Shape reconstruction (object) | F1 @2cm (%): 33.64 -> 90.42 | Improved |
| use RHINO instead of HSR | RHINO · Shape reconstruction (human) | Chamfer distance (cm): 2.69 -> 2.65 | No clear effect |
| use RHINO instead of HSR | RHINO · Novel-view synthesis | PSNR: 22.65 -> 25.80 | Improved |
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.
- HOLD — HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024) · better on 5 of 5 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
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.90)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.89)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.89)
- PICO: Reconstructing 3D People In Contact with Objects (CVPR 2025 · similarity 0.89)
- Real2Sim in HOI: Toward Physically Plausible HOI Reconstruction from Monocular Videos (NeurIPS 2026 · similarity 0.88)
- Reconstructing In-the-Wild Open-Vocabulary Human-Object Interactions (CVPR 2025 · 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.
