Research map / 3D vision / Face and shape
DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 projected local image features to global features (ablation) | DISN · ShapeNet chair | EMD: 2.75 -> 2.62 | Helped |
| add projected local image features (ablation) | DISN · ShapeNet chair | IoU: 54.8 -> 55.3 | Helped |
| regress continuous SDF instead of binary occupancy classification (ablation) | DISN · ShapeNet chair | CD: 8.27 -> 7.55 | Helped |
| two-stream decoders instead of one-stream (ablation) | DISN · ShapeNet chair | EMD: 2.71 -> 2.62 | Helped |
| use ground-truth camera instead of estimated camera (ablation) | DISN · ShapeNet Core | IoU: 57.0 -> 59.4 | Helped |
| DISN vs implicit baselines OccNet/IMNET | single-view reconstruction · ShapeNet Core | IoU: 56.4 -> 59.4 | Helped |
| DISN vs 3DN | single-view reconstruction · ShapeNet Core | CD: 9.77 vs 10.17 | Hurt |
| 6D rotation representation with point-cloud pose loss | camera pose estimation · ShapeNet | 2D reprojection error: 4.86 -> 2.95 | Helped |
Closest papers by meaning
- D2IM-Net: Learning Detail Disentangled Implicit Fields From Single Images (CVPR 2021 · similarity 0.91)
- Implicit Functions in Feature Space for 3D Shape Reconstruction and Completion (CVPR 2020 · similarity 0.88)
- MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image (AAAI 2019 · similarity 0.87)
- NeAT: Learning Neural Implicit Surfaces With Arbitrary Topologies From Multi-View Images (CVPR 2023 · similarity 0.87)
- Learning to Infer Implicit Surfaces without 3D Supervision (NeurIPS 2019 · similarity 0.87)
- SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration (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.