Research map / 3D vision / 3D generation
Learning Relationships for Multi-View 3D Object Recognition
Ze Yang, Liwei 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 Integrating block instead of view pooling (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 93.8 -> 94.3 | Improved |
| add a Reinforcing block after conv5 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 93.7 -> 94.3 | Improved |
| place the Reinforcing block after conv5 instead of conv2 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 93.9 -> 94.3 | Improved |
| place the Reinforcing block after conv5 instead of conv3 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 93.7 -> 94.3 | Improved |
| place the Reinforcing block after conv5 instead of conv4 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 94.0 -> 94.3 | Improved |
| add a second Reinforcing block after conv4 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 94.3 -> 93.9 | No clear effect |
| add Reinforcing blocks after conv3,4,5 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 94.3 -> 94.2 | No clear effect |
| add Reinforcing blocks after conv2,3,4,5 (ablation) | Relation Network · ModelNet40 | instance accuracy (%): 94.3 -> 93.8 | Hurt |
| add low-rank Mahalanobis metric learning (ablation) | Relation Network · ModelNet40 retrieval | mAP: 82.7 -> 86.7 | Improved |
| increase number of views from 6 to 12 | Relation Network · ModelNet40 | instance accuracy (%): 94.1 -> 94.3 | Improved |
| increase number of views from 6 to 12 | MHBN · ModelNet40 | instance accuracy (%): 94.1 -> 93.4 | Hurt |
| increase number of views from 6 to 12 | MVCNN · ModelNet40 | instance accuracy (%): 92.0 -> 91.5 | Hurt |
Closest papers by meaning
- View N-Gram Network for 3D Object Retrieval (ICCV 2019 · similarity 0.89)
- Beyond Viewpoint: Robust 3D Object Recognition under Arbitrary Views through Joint Multi-Part Representation (ECCV 2024 · similarity 0.88)
- Pairwise Decomposition of Image Sequences for Active Multi-View Recognition (CVPR 2016 · similarity 0.87)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.87)
- Multiview Scene Graph (NeurIPS 2024 · similarity 0.85)
- Object-Centric Representation Learning for Enhanced 3D Semantic Scene Graph Prediction (NeurIPS 2025 · similarity 0.85)
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.
