Research map / 3D vision / Face and shape
2D Amodal Instance Segmentation Guided by 3D Shape Prior
Zhixuan Li, Weining Ye, Tingting Jiang, Tiejun Huang
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 ShapeNet and PASCAL3D+ 3D-reconstruction pretraining (ablation) | A3D · D2SA | mAP: 68.5 -> 73.5 | Improved |
| add ShapeNet and PASCAL3D+ pretraining (ablation) | A3D · KINS | mAP: 31.4 -> 36.2 | Improved |
| add ShapeNet and PASCAL3D+ pretraining (ablation) | A3D · COCOA-cls | mAP: 34.9 -> 40.6 | Improved |
| add same pretraining data to ShapeDict | ShapeDict · D2SA | mAP: 70.3 -> 71.2 | No clear effect |
| add pretraining on PASCAL3D+ instead of ShapeNet only (ablation) | A3D · COCOA-cls | mAP: 37.4 -> 39.2 | Improved |
| add Category-specific 3D Modeling module to Mask-RCNN baseline (ablation) | Mask-RCNN · ShapeNet | mIoU: 4% mIoU improvement | Improved |
| add Region-specific Edge Refine module (ablation) | A3D · ShapeNet | mIoU: 0.7% mIoU | Improved |
| add cross-view technique in pretraining (ablation) | A3D · ShapeNet | mIoU: 82.5 -> 83.9 | Improved |
| add 3D shape prior without pretraining | A3D vs ShapeDict · D2SA | mAP: 70.3 -> 68.5 | Hurt |
| fixed-topology sphere mesh on complex-structure categories | A3D · ShapeNet Chair | mIoU: drops 9.6% and 11.7% IoU on Chair and Lamp | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- BCNet — BCNet: Learning Body and Cloth Shape from A Single Image (ECCV 2020) · better on 1 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
- Amodal Ground Truth and Completion in the Wild (CVPR 2024 · similarity 0.91)
- BLADE: Box-Level Supervised Amodal Segmentation through Directed Expansion (AAAI 2024 · similarity 0.89)
- Segment Anything, Even Occluded (CVPR 2025 · similarity 0.89)
- pix2gestalt: Amodal Segmentation by Synthesizing Wholes (CVPR 2024 · similarity 0.88)
- Using Diffusion Priors for Video Amodal Segmentation (CVPR 2025 · similarity 0.88)
- Amodal Scene Analysis via Holistic Occlusion Relation Inference and Generative Mask Completion (AAAI 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.
