Research map / 3D vision / Face and shape
DANBO: Disentangled Articulated Neural Body Representations via Graph Neural Networks
Shih-Yang Su, Timur Bagautdinov, Helge Rhodin
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 learned aggregation network (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 17.08 -> 24.38 | Improved |
| add factorized per-bone volumes (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 24.24 -> 24.38 | Improved |
| add GNN features (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 23.87 -> 24.38 | Improved |
| use Soft-softmax instead of softmax aggregation (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 23.80 -> 24.38 | Improved |
| use Soft-softmax instead of Softmax-OOB (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 24.00 -> 24.38 | Improved |
| use weighted aggregation instead of unweighted Sum-OOB (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 23.22 -> 24.38 | Improved |
| use factorized volume instead of SlabConv full 3D volume (ablation) | DANBO · Human3.6M S9 ablation | PSNR: 24.17 -> 24.38 | Improved |
| DANBO versus A-NeRF in novel view synthesis | A-NeRF baseline · Human3.6M novel view | PSNR: 24.21 -> 24.70 | Improved |
| DANBO versus A-NeRF in novel pose synthesis | A-NeRF baseline · Human3.6M novel pose | PSNR: 22.96 -> 23.76 | Improved |
| DANBO versus A-NeRF on MonoPerfCap | A-NeRF baseline · MonoPerfCap novel pose | PSNR: 23.67 -> 23.25 | Mixed |
Later papers that compared with this method
3 later accepted papers report DANBO as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Pose Modulated Avatars from Video (ICLR 2024 · better on 10 of 10 shared results)
- Gaussian Shadow Casting for Neural Characters (CVPR 2024 · better on 8 of 8 shared results)
- NPC: Neural Point Characters from Video (ICCV 2023 · better on 5 of 6 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.
- NeuralBody — Neural Body: Implicit Neural Representations With Structured Latent Codes for Novel View Synthesis of Dynamic Humans (CVPR 2021) · better on 4 of 4 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
- Pose Modulated Avatars from Video (ICLR 2024 · similarity 0.88)
- BANMo: Building Animatable 3D Neural Models From Many Casual Videos (CVPR 2022 · similarity 0.88)
- Textured Neural Avatars (CVPR 2019 · similarity 0.88)
- MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Images (NeurIPS 2021 · similarity 0.87)
- Neural Pose Representation Learning for Generating and Transferring Non-Rigid Object Poses (NeurIPS 2024 · similarity 0.86)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (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.
