Research map / 3D vision / Face and shape
C3DPO: Canonical 3D Pose Networks for Non-Rigid Structure From Motion
David Novotny, Nikhila Ravi, Benjamin Graham, Natalia Neverova, Andrea Vedaldi
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 canonicalization network and in-plane rotation loss to C3DPO-base (ablation) | C3DPO · S-Up3D | MPJPE: 0.160 -> 0.068 | Improved |
| add canonicalization network on top of C3DPO-equiv (ablation) | C3DPO · S-Up3D | MPJPE: 0.154 -> 0.068 | Improved |
| add in-plane rotation invariant loss to C3DPO-base (ablation) | C3DPO-base · S-Up3D | MPJPE: 0.160 -> 0.154 | Improved |
| add canonicalization network on top of C3DPO-equiv (ablation) | C3DPO · PASCAL3D+ (12 classes, GT keypoints at test) | MPJPE: 50.1 -> 38.0 | Improved |
| add in-plane rotation invariant loss to C3DPO-base (ablation) | C3DPO-base · PASCAL3D+ (12 classes, GT keypoints at test) | MPJPE: 53.5 -> 50.1 | Improved |
| add canonicalization network on top of C3DPO-equiv (ablation) | C3DPO · Human3.6M (ground truth 2D keypoints) | MPJPE: 128.2 -> 101.8 | Improved |
| add canonicalization network on top of C3DPO-equiv (ablation) | C3DPO · Human3.6M (Stacked Hourglass keypoints) | MPJPE: 190.4 -> 153.0 | Improved |
| add in-plane rotation invariant loss to C3DPO-base (ablation) | C3DPO-base · Human3.6M (Stacked Hourglass keypoints) | MPJPE: 201.6 -> 190.4 | Improved |
| use C3DPO-base instead of Pose-GAN with ground truth keypoints (re-projection loss only) | C3DPO-base · Human3.6M (ground truth 2D keypoints) | MPJPE: 130.9 -> 135.2 | Hurt |
| use C3DPO + HRNet instead of CMR | C3DPO · PASCAL3D+ (12 classes, predicted keypoints) | MPJPE: 74.4 -> 57.5 | Improved |
| use C3DPO instead of GbNrSfM | NR-SFM baseline · PASCAL3D+ (12 classes, GT keypoints at test) | MPJPE: 184.6 -> 38.0 | Improved |
| use C3DPO instead of EM-SfM on S-Up3D | NR-SFM baseline · S-Up3D | MPJPE: 0.107 -> 0.068 | Improved |
Later papers that compared with this method
6 later accepted papers report C3dpo as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.
- Deep Non-Rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling (AAAI 2025 · better on 3 of 3 shared results)
- Deep Non-Rigid Structure-From-Motion: A Sequence-to-Sequence Translation Perspective (TPAMI 2024 · better on 8 of 8 shared results)
- MHR-Net: Multiple-Hypothesis Reconstruction of Non-rigid Shapes from 2D Views (ECCV 2022 · better on 4 of 4 shared results)
- DensePose 3D: Lifting Canonical Surface Maps of Articulated Objects to the Third Dimension (ICCV 2021 · better on 4 of 4 shared results)
- PAUL: Procrustean Autoencoder for Unsupervised Lifting (CVPR 2021 · better on 1 of 1 shared result)
- Procrustean Regression Networks: Learning 3D Structure of Non-Rigid Objects from 2D Annotations (ECCV 2020 · 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
- BANMo: Building Animatable 3D Neural Models From Many Casual Videos (CVPR 2022 · similarity 0.88)
- One-shot 3D Object Canonicalization based on Geometric and Semantic Consistency (CVPR 2025 · similarity 0.87)
- Unsupervised Learning of Category-Level 3D Pose from Object-Centric Videos (CVPR 2024 · similarity 0.87)
- Deep Non-Rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling (AAAI 2025 · similarity 0.87)
- DualPM: Dual Posed-Canonical Point Maps for 3D Shape and Pose Reconstruction (CVPR 2025 · similarity 0.86)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 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.
