Research map / 3D vision / 3D generation
Volumetric and Multi-View CNNs for Object Classification on 3D Data
Check 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 multi-resolution 3D sphere-rendered features (ablation) | MVCNN · ModelNet40 | classification accuracy (instance average): 92.0 -> 93.8 | Helped |
| Add multi-resolution 3D sphere-rendered features (ablation) | MVCNN · ModelNet40 | classification accuracy (class average): 89.7 -> 91.4 | Helped |
| Add multi-resolution 3D sphere-rendered features | MVCNN · Real-world reconstructions | retrieval MAP: 35.8 -> 51.4 | Helped |
| Add auxiliary subvolume classification supervision | 3D-NIN volumetric CNN · ModelNet40 | classification accuracy (instance average, Multi-Ori): 88.5 -> 89.2 | Helped |
| Add anisotropic probing kernels | 3D-NIN volumetric CNN · ModelNet40 | classification accuracy (instance average, Multi-Ori): 88.5 -> 89.9 | Helped |
| Add multi-orientation pooling with azimuth rotation (ablation) | Volumetric CNN · ModelNet40 | classification accuracy: 84.7 -> 86.1 | Helped |
| Add multi-orientation pooling with azimuth and elevation rotation (ablation) | Volumetric CNN · ModelNet40 | classification accuracy: 83.0 -> 87.8 | Helped |
| Add translation jittering to azimuth augmentation and multi-orientation pooling (ablation) | Volumetric CNN · ModelNet40 | classification accuracy: 84.8 -> 86.1 | No clear effect |
| Add multi-orientation pooling (ablation) | E2E volumetric CNN · ModelNet40 | classification accuracy: 83.0 -> 87.8 | Helped |
| Add multi-orientation pooling (ablation) | VoxNet · ModelNet40 | classification accuracy: 83.8 -> 85.9 | Helped |
| Add multi-orientation pooling (ablation) | 3D-NIN · ModelNet40 | classification accuracy: 86.1 -> 88.5 | Helped |
| Add multi-orientation pooling (ablation) | Ours-SubvolumeSup · ModelNet40 | classification accuracy: 87.2 -> 89.2 | Helped |
Closest papers by meaning
- FPNN: Field Probing Neural Networks for 3D Data (NeurIPS 2016 · similarity 0.88)
- Pixels, Voxels, and Views: A Study of Shape Representations for Single View 3D Object Shape Prediction (CVPR 2018 · similarity 0.88)
- A Comparative Analysis and Study of Multiview CNN Models for Joint Object Categorization and Pose Estimation (ICML 2016 · similarity 0.87)
- OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentation (CVPR 2024 · similarity 0.87)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.86)
- Duoduo CLIP: Efficient 3D Understanding with Multi-View Images (ICLR 2025 · 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.