Research map / 3D vision / Face and shape
ViSER: Video-Specific Surface Embeddings for Articulated 3D Shape Reconstruction
Gengshan Yang, Deqing Sun, Varun Jampani, Daniel Vlasic, Forrester Cole, Ce Liu, Deva Ramanan
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 3D matching loss (ablation) | ViSER · break-1 | keypoint transfer accuracy: 36.2 -> 70.5 | Improved |
| add 3D matching loss (ablation) | ViSER · Elephants inner | keypoint transfer accuracy: 51.3 -> 80.4 | Improved |
| add 3D matching loss (ablation) | ViSER · Elephants cross | keypoint transfer accuracy: 42.6 -> 68.9 | Improved |
| add 2D cycle reprojection loss (ablation) | ViSER · break-1 | keypoint transfer accuracy: 38.3 -> 70.5 | Improved |
| add 2D cycle reprojection loss (ablation) | ViSER · Elephants cross | keypoint transfer accuracy: 62.5 -> 68.9 | Improved |
| add 2D cycle reprojection loss (ablation) | ViSER · Elephants inner | keypoint transfer accuracy: 80.1 -> 80.4 | No clear effect |
| use pixel-surface embedding matching instead of CSM regression (ablation) | ViSER · break-1 | keypoint transfer accuracy: 47.1 -> 70.5 | Improved |
| use pixel-surface embedding matching instead of CSM regression (ablation) | ViSER · Elephants inner | keypoint transfer accuracy: 77.4 -> 80.4 | Improved |
| use ViSER instead of template-based methods on tight-clothing common poses | ViSER vs DensePose CSE · parkour | keypoint transfer accuracy: 85.9 -> 62.9 | Hurt |
Later papers that compared with this method
4 later accepted papers report ViSER as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- S3O: A Dual-Phase Approach for Reconstructing Dynamic Shape and Skeleton of Articulated Objects from Single Monocular Video (ICML 2024 · better on 4 of 4 shared results)
- Root Pose Decomposition Towards Generic Non-rigid 3D Reconstruction with Monocular Videos (ICCV 2023 · better on 1 of 1 shared result)
- BANMo: Building Animatable 3D Neural Models From Many Casual Videos (CVPR 2022 · better on 6 of 6 shared results)
- CASA: Category-agnostic Skeletal Animal Reconstruction (NeurIPS 2022 · better on 5 of 5 shared results)
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.
- LASR — LASR: Learning Articulated Shape Reconstruction From a Monocular Video (CVPR 2021) · better on 5 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
- Learning Monocular 3D Reconstruction of Articulated Categories From Motion (CVPR 2021 · similarity 0.89)
- LASR: Learning Articulated Shape Reconstruction From a Monocular Video (CVPR 2021 · similarity 0.89)
- KeyTr: Keypoint Transporter for 3D Reconstruction of Deformable Objects in Videos (CVPR 2022 · similarity 0.88)
- Learning Implicit Representation for Reconstructing Articulated Objects (ICLR 2024 · similarity 0.88)
- S3O: A Dual-Phase Approach for Reconstructing Dynamic Shape and Skeleton of Articulated Objects from Single Monocular Video (ICML 2024 · similarity 0.87)
- MeshPose: Unifying DensePose and 3D Body Mesh Reconstruction (CVPR 2024 · similarity 0.86)
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.
