Research map / 3D vision / Face and shape
SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Images
Chen-Hsuan Lin, Chaoyang Wang, Simon Lucey
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 |
|---|---|---|---|
| Use SDF-SRN instead of SoftRas | ShapeNet airplane reconstruction with unassociated views · ShapeNet airplane | accuracy (×10; lower is better): 0.250 -> 0.193 | Improved |
| Use SDF-SRN instead of SoftRas | ShapeNet car reconstruction with unassociated views · ShapeNet car | accuracy (×10; lower is better): 0.356 -> 0.141 | Improved |
| Use SDF-SRN instead of SoftRas | ShapeNet chair reconstruction with unassociated views · ShapeNet chair | accuracy (×10; lower is better): 0.690 -> 0.352 | Improved |
| Use SDF-SRN instead of CMR | PASCAL3D+ airplane reconstruction · PASCAL3D+ airplane | accuracy (×10; lower is better): 0.625 -> 0.582 | Improved |
| Use SDF-SRN instead of CMR | PASCAL3D+ car reconstruction · PASCAL3D+ car | accuracy (×10; lower is better): 0.474 -> 0.391 | Improved |
| Use SDF-SRN instead of CMR | PASCAL3D+ chair reconstruction · PASCAL3D+ chair | accuracy (×10; lower is better): 1.396 -> 0.478 | Improved |
| Use an SDF objective rather than binary occupancy classification (ablation) | SDF-SRN · ShapeNet chairs | accuracy (×10; lower is better): 0.425 -> 0.352 | Improved |
| Add rendering loss (L_RGB) (ablation) | SDF-SRN · ShapeNet chairs | coverage (×10; lower is better): 0.554 -> 0.315 | Improved |
| Add importance weighting (ablation) | SDF-SRN · ShapeNet chairs | coverage (×10; lower is better): 1.168 -> 0.315 | Improved |
| Add positional encoding (ablation) | SDF-SRN · ShapeNet chairs | accuracy (×10; lower is better): 0.444 -> 0.352 | Improved |
| Optimize the latent code at test time with known camera pose (ablation) | SDF-SRN · ShapeNet chairs | accuracy (×10; lower is better): 0.352 -> 0.332 | Improved |
Later papers that compared with this method
3 later accepted papers report SDF-SRN as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Topologically-Aware Deformation Fields for Single-View 3D Reconstruction (CVPR 2022 · better on 6 of 9 shared results)
- 2D GANs Meet Unsupervised Single-View 3D Reconstruction (ECCV 2022 · better on 4 of 4 shared results)
- Planes vs. Chairs: Category-Guided 3D Shape Learning without Any 3D Cues (ECCV 2022 · better on 2 of 2 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
- SDFDiff: Differentiable Rendering of Signed Distance Fields for 3D Shape Optimization (CVPR 2020 · similarity 0.90)
- Learning Scene-Level Signed Directional Distance Function With Ellipsoidal Priors and Neural Residuals (TPAMI 2026 · similarity 0.90)
- A-SDF: Learning Disentangled Signed Distance Functions for Articulated Shape Representation (ICCV 2021 · similarity 0.89)
- Few-Shot Unsupervised Implicit Neural Shape Representation Learning with Spatial Adversaries (ICML 2024 · similarity 0.89)
- Learning Compact Latent Space for Representing Neural Signed Distance Functions with High-fidelity Geometry Details (AAAI 2026 · similarity 0.89)
- DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (CVPR 2019 · similarity 0.89)
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.
