Research map / 3D vision / Face and shape
Emphasizing 3D Properties in Recurrent Multi-View Aggregation for 3D Shape Retrieval
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 |
|---|---|---|---|
| aggregate views with LSTM plus viewpoint-estimation decoder instead of max pooling (CNN+Max) | ERFA-Net · ModelNet40 | MAP: 79.00% -> 85.38% | Helped |
| aggregate views with LSTM plus decoder instead of concatenation (CNN+Concat) | ERFA-Net · ModelNet40 | MAP: 80.49% -> 85.38% | Helped |
| aggregate views with LSTM plus decoder instead of concatenation (CNN+Concat) | ERFA-Net · ModelNet10 | MAP: 91.57% -> 93.24% | Helped |
| aggregate views with LSTM plus decoder instead of max pooling (CNN+Max) | ERFA-Net · ModelNet10 | MAP: 89.55% -> 93.24% | Helped |
| use ERFA-Net instead of MVCNN | ERFA-Net · SHREC16 Perturbed (Micro+Macro) | MAP: 0.698 -> 0.857 | Helped |
| use ERFA-Net instead of MVCNN | ERFA-Net · SHREC16 Normal (Micro+Macro) | MAP: 0.845 -> 0.861 | Helped |
| use ERFA-Net instead of MVCNN on Macro F-measure | ERFA-Net · SHREC16 Normal (Macro) | F-measure: 0.575 -> 0.574 | No clear effect |
| train with generative view-estimation supervision, test with 6 of 12 views replaced by noise | ERFA-Net · ModelNet40 noisy views | MAP: 81.15% | Helped |
Closest papers by meaning
- Multiview Aggregation for Learning Category-Specific Shape Reconstruction (NeurIPS 2019 · similarity 0.89)
- MLVCNN: Multi-Loop-View Convolutional Neural Network for 3D Shape Retrieval (AAAI 2019 · similarity 0.87)
- Angular Triplet-Center Loss for Multi-View 3D Shape Retrieval (AAAI 2019 · similarity 0.87)
- View-GCN: View-Based Graph Convolutional Network for 3D Shape Analysis (CVPR 2020 · similarity 0.87)
- GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition (CVPR 2018 · similarity 0.86)
- Single-View 3D Object Reconstruction From Shape Priors in Memory (CVPR 2021 · 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.