Research map / Point clouds / Segmentation and detection
ImVoteNet: Boosting 3D Object Detection in Point Clouds With Image Votes
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 lifted image votes and image cues to VoteNet | VoteNet · SUN RGB-D v1 val | mAP@0.25: 57.7 -> 63.4; +5.7 mAP | Helped |
| append raw RGB values to point features (ablation) | VoteNet · SUN RGB-D v1 val | mAP@0.25: 57.7 -> 56.3 | Hurt |
| add one-hot region class-confidence features (ablation) | VoteNet · SUN RGB-D v1 val | mAP@0.25: 57.7 -> 59.6 | Helped |
| add the lifted 2D vote and ray angle (ablation) | ImVoteNet · SUN RGB-D | mAP: 61.2 -> 63.4 | Helped |
| add ray angle (ablation) | ImVoteNet with 2D vote · SUN RGB-D | mAP: 62.2 -> 63.4 | Helped |
| use one-hot class score instead of 64-d RoI features (ablation) | ImVoteNet · SUN RGB-D | mAP: 62.4 -> 63.4 | Helped |
| use one-hot class score instead of 1024-d RoI features (ablation) | ImVoteNet · SUN RGB-D | mAP: 59.5 -> 63.4 | Helped |
| add region semantic cues (ablation) | ImVoteNet · SUN RGB-D | mAP: 58.9 -> 63.4 | Helped |
| use raw RGB instead of FPN-P2 or FPN-P3 pixel features (ablation) | ImVoteNet · SUN RGB-D | mAP: 62.0 -> 63.4 | Helped |
| add raw RGB texture cue (ablation) | ImVoteNet · SUN RGB-D | mAP: 62.4 -> 63.4 | Helped |
| use balanced gradient-blending tower weights (ablation) | ImVoteNet · SUN RGB-D joint tower | mAP: 62.1 -> 63.4 | Helped |
| add image cues to point-cloud detection | VoteNet · SUN RGB-D, ORB 1k points | mAP: 27.9 -> 47.1; +19.2 | Helped |
Closest papers by meaning
- Prototypical VoteNet for Few-Shot 3D Point Cloud Object Detection (NeurIPS 2022 · similarity 0.90)
- ImOV3D: Learning Open Vocabulary Point Clouds 3D Object Detection from Only 2D Images (NeurIPS 2024 · similarity 0.89)
- Back-Tracing Representative Points for Voting-Based 3D Object Detection in Point Clouds (CVPR 2021 · similarity 0.88)
- Point2Seq: Detecting 3D Objects As Sequences (CVPR 2022 · similarity 0.88)
- MLCVNet: Multi-Level Context VoteNet for 3D Object Detection (CVPR 2020 · similarity 0.88)
- SPGroup3D: Superpoint Grouping Network for Indoor 3D Object Detection (AAAI 2024 · 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.