Research map / 3D vision / Face and shape
A-SDF: Learning Disentangled Signed Distance Functions for Articulated Shape Representation
Jiteng Mu, Weichao Qiu, Adam Kortylewski, Alan Yuille, Nuno Vasconcelos, Xiaolong Wang
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 test-time adaptation of the shape encoder (ablation) | A-SDF · Shape2Motion Oven reconstruction | Chamfer-L1 (x1000): 3.72 -> 1.83 | Improved |
| add test-time adaptation of the shape encoder (ablation) | A-SDF · Shape2Motion Fridge reconstruction | Chamfer-L1 (x1000): 0.90 -> 0.69 | Improved |
| add test-time adaptation of the shape encoder (ablation) | A-SDF · RBO real depth reconstruction | Chamfer-L1 (x1000): 2.53 -> 0.76 | Improved |
| add test-time adaptation of the shape encoder (ablation) | A-SDF · RBO real depth generation | Chamfer-L1 (x1000): 5.09 -> 3.22 | Improved |
| add test-time adaptation of the shape encoder (ablation) | A-SDF · Laptop partial point cloud 1-view reconstruction | Chamfer-L1 (x1000): 2.19 -> 2.25 | Hurt |
| add test-time adaptation of the shape encoder (ablation) | A-SDF · Laptop partial point cloud 2-view reconstruction | Chamfer-L1 (x1000): 0.75 -> 0.61 | Improved |
| add test-time adaptation of the shape encoder (ablation) | A-SDF · Shape2Motion shape synthesis (7 categories) | Chamfer-L1 (x1000): 2.86 -> 3.01 | Mixed |
| add part label supervision (ablation) | A-SDF with TTA · Shape2Motion Washing machine shape synthesis | Chamfer-L1 (x1000): 3.01 -> 1.96 | Improved |
| add part label supervision (ablation) | A-SDF with TTA · Shape2Motion Oven shape synthesis | Chamfer-L1 (x1000): 2.58 -> 3.10 | Hurt |
| use disentangled shape and articulation codes instead of single DeepSDF code | A-SDF w/o TTA · Shape2Motion Oven interpolation | Chamfer-L1 (x1000): 11.13 -> 6.83 | Improved |
| use disentangled shape and articulation codes instead of single DeepSDF code | A-SDF w/o TTA · Shape2Motion Laptop interpolation | Chamfer-L1 (x1000): 2.77 -> 0.17 | Improved |
| use disentangled codes instead of single DeepSDF code | A-SDF · Shape2Motion Laptop extrapolation | Chamfer-L1 (x1000): 3.396 -> 0.280 | Improved |
Later papers that compared with this method
5 later accepted papers report A-SDF as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- SIAM: Towards Generalizable Articulated Object Modeling via Single Robot-Object Interaction (AAAI 2026 · better on 1 of 1 shared result)
- CARTO: Category and Joint Agnostic Reconstruction of ARTiculated Objects (CVPR 2023 · better on 5 of 7 shared results)
- PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects (ICCV 2023 · better on 1 of 1 shared result)
- CaDeX: Learning Canonical Deformation Coordinate Space for Dynamic Surface Representation via Neural Homeomorphism (CVPR 2022 · better on 2 of 2 shared results)
- Ditto: Building Digital Twins of Articulated Objects From Interaction (CVPR 2022 · better on 1 of 1 shared result)
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.
- DeepSDF — DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (CVPR 2019) · better on 11 of 11 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
- DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (CVPR 2019 · similarity 0.90)
- Learning Compact Latent Space for Representing Neural Signed Distance Functions with High-fidelity Geometry Details (AAAI 2026 · similarity 0.89)
- SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Images (NeurIPS 2020 · similarity 0.89)
- Few-Shot Unsupervised Implicit Neural Shape Representation Learning with Spatial Adversaries (ICML 2024 · similarity 0.88)
- gSDF: Geometry-Driven Signed Distance Functions for 3D Hand-Object Reconstruction (CVPR 2023 · similarity 0.88)
- Small Steps and Level Sets: Fitting Neural Surface Models with Point Guidance (CVPR 2024 · similarity 0.87)
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.
