Research map / 3D vision / Face and shape
GSIR: Generalizable 3D Shape Interpretation and Reconstruction
Jianren Wang, Zhaoyuan Fang
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 interpretation consistency refinement to structure recovery (ablation) | GSIR · ShapeNet structure recovery | Hausdorff error: 0.0798 -> 0.0731 | Improved |
| add interpretation consistency refinement to structure recovery (ablation) | GSIR · ShapeNet structure recovery | thresholded accuracy (delta<0.1): 79.6% -> 84.8% | Improved |
| add consistency loss to reconstruction (seen) (ablation) | GSIR · ShapeNet reconstruction (seen categories) | Chamfer distance: 0.060 -> 0.057 | Improved |
| add consistency loss to reconstruction (unseen) (ablation) | GSIR · ShapeNet reconstruction (unseen categories) | Chamfer distance: 0.103 -> 0.099 | Improved |
| add depth estimation module to encoder-decoder (unseen) (ablation) | Encoder Decoder · ShapeNet reconstruction (unseen categories) | Chamfer distance: 0.196 -> 0.131 | Improved |
| replace multi-view with spherical map inpainting (unseen) (ablation) | Depth + Vi · ShapeNet reconstruction (unseen categories) | Chamfer distance: 0.123 -> 0.119 | Improved |
| replace multi-view with spherical map inpainting (seen) (ablation) | Depth + Vi · ShapeNet reconstruction (seen categories) | Chamfer distance: 0.075 -> 0.073 | No clear effect |
| add structure guidance to spherical map inpainting (seen) (ablation) | Depth + Spherical Map + Vi · ShapeNet reconstruction (seen categories) | Chamfer distance: 0.073 -> 0.069 | Improved |
| add structure guidance to spherical map inpainting (unseen) (ablation) | Depth + Spherical Map + Vi · ShapeNet reconstruction (unseen categories) | Chamfer distance: 0.119 -> 0.113 | Improved |
| add structure guidance to multi-view inpainting (seen) (ablation) | Depth + Multi-view + Vi · ShapeNet reconstruction (seen categories) | Chamfer distance: 0.075 -> 0.072 | Improved |
| add partial voxel grid Vp to refinement input (seen) (ablation) | Depth + Spherical Map + Guided + Vi · ShapeNet reconstruction (seen categories) | Chamfer distance: 0.069 -> 0.064 | Improved |
| add partial voxel grid Vp to refinement input (unseen) (ablation) | Depth + Spherical Map + Guided + Vi · ShapeNet reconstruction (unseen categories) | Chamfer distance: 0.113 -> 0.106 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- DeepSDF — DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (CVPR 2019) · better on 1 of 2 shared results
- MarrNet — MarrNet: 3D Shape Reconstruction via 2.5D Sketches (NeurIPS 2017) · 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
- Learning to Reconstruct Shapes from Unseen Classes (NeurIPS 2018 · similarity 0.89)
- Fostering Generalization in Single-View 3D Reconstruction by Learning a Hierarchy of Local and Global Shape Priors (CVPR 2021 · similarity 0.89)
- Few-Shot Generalization for Single-Image 3D Reconstruction via Priors (ICCV 2019 · similarity 0.88)
- Deep Optimized Priors for 3D Shape Modeling and Reconstruction (CVPR 2021 · similarity 0.87)
- Front2Back: Single View 3D Shape Reconstruction via Front to Back Prediction (CVPR 2020 · similarity 0.87)
- GSNet: Joint Vehicle Pose and Shape Reconstruction with Geometrical and Scene-aware Supervision (ECCV 2020 · 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.
