Research map / Detection and segmentation / 3D detection and driving
Multi-View Harmonized Bilinear Network for 3D Object Recognition
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 learned singular-value harmonization instead of fixed square-root normalization | Bilinear pooling for 3D object recognition · ModelNet40 | average instance accuracy: 93.23 -> 94.12 | Helped |
| Add learned singular-value harmonization instead of fixed square-root normalization | Bilinear pooling for 3D object recognition · ModelNet10 | average instance accuracy: 93.17 -> 94.93 | Helped |
| Add early sqrt and late sqrt sub-layers (ablation) | MHBN · ModelNet40 | average instance accuracy: 93.24 -> 94.12 | Helped |
| Add late sqrt sub-layer (ablation) | MHBN · ModelNet40 | average instance accuracy: 93.24 -> 93.63 | Helped |
| Add early sqrt and late sqrt sub-layers (ablation) | MHBN · ModelNet10 | average instance accuracy: 93.39 -> 94.93 | Helped |
| Add early sqrt and late sqrt sub-layers (ablation) | MHBN · ModelNet40 | average class accuracy: 90.00 -> 92.23 | Helped |
| Use 12 views instead of 6 (ablation) | MHBN · ModelNet40 | average instance accuracy: 94.12 -> 93.42 | Hurt |
| Use 3 views instead of 6 (ablation) | MHBN · ModelNet40 | average instance accuracy: 94.12 -> 93.78 | Hurt |
| Add depth-view features to RGB-view features | MHBN · ModelNet40 | average instance accuracy: 94.1 -> 94.7 | Helped |
| Add depth-view features to RGB-view features | MHBN · ModelNet10 | average instance accuracy: 94.9 -> 95.0 | No clear effect |
| Increase local feature dimension above 64 (ablation) | MHBN · ModelNet10 | accuracy | Hurt |
| Increase local feature dimension (ablation) | MHBN · ModelNet40 | accuracy | Helped |
Closest papers by meaning
- Compact Bilinear Pooling (CVPR 2016 · similarity 0.88)
- Efficient Compact Bilinear Pooling via Kronecker Product (AAAI 2022 · similarity 0.85)
- Fast and Compact Bilinear Pooling by Shifted Random Maclaurin (AAAI 2021 · similarity 0.85)
- Revisiting Bilinear Pooling: A Coding Perspective (AAAI 2020 · similarity 0.84)
- Multi-view Aggregation Network for Dichotomous Image Segmentation (CVPR 2024 · similarity 0.84)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.83)
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.