Research map / Detection and segmentation / 3D detection and driving
Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspots
Qi Chen, Lin Sun, Zhixin Wang, Kui Jia, Alan Yuille
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 quadrant spatial relation encoder (ablation) | HotSpotNet · KITTI val Cyclist Moderate | AP: 69.31 -> 72.55 | Improved |
| add quadrant spatial relation encoder (ablation) | HotSpotNet · KITTI val Car Moderate | AP: 82.27 -> 82.75 | Improved |
| add quadrant spatial relation encoder (ablation) | HotSpotNet · KITTI val Cyclist Easy | AP: 89.48 -> 88.22 | Hurt |
| add soft argmin for regression (ablation) | HotSpotNet · KITTI val Cyclist Moderate | AP: 68.65 -> 72.55 | Improved |
| add soft argmin for regression (ablation) | HotSpotNet · KITTI val Pedestrian Moderate | AP: 63.7 -> 65.9 | Improved |
| limit hotspots per object to C=64 instead of all non-empty spots (ablation) | HotSpotNet · KITTI val Cyclist Moderate | AP: 67.66 -> 72.55 | Improved |
| limit hotspots per object to C=64 instead of all non-empty spots (ablation) | HotSpotNet · KITTI val Pedestrian Moderate | AP: 62.08 -> 65.9 | Improved |
| limit hotspots per object to C=64 instead of all non-empty spots (ablation) | HotSpotNet · KITTI val Car Moderate | AP: 82.93 -> 82.75 | No clear effect |
| assign all non-empty spots rather than all voxels as hotspots (ablation) | HotSpotNet · KITTI val Pedestrian Moderate | AP: 62.82 -> 62.08 | No clear effect |
| use eight-direction spatial encoding instead of quadrants (ablation) | HotSpotNet · KITTI val Cyclist Moderate | AP: 72.55 -> 69.99 | Hurt |
| use left and right encoding instead of none (ablation) | HotSpotNet · KITTI val Pedestrian Moderate | AP: 65.45 -> 64.05 | Hurt |
| use deviation regression instead of quadrant classification (ablation) | HotSpotNet · KITTI val Car Moderate | AP: 82.75 -> 82.03 | Hurt |
Closest papers by meaning
- PillarNeXt: Rethinking Network Designs for 3D Object Detection in LiDAR Point Clouds (CVPR 2023 · similarity 0.91)
- Approaching Outside: Scaling Unsupervised 3D Object Detection from 2D Scene (ECCV 2024 · similarity 0.90)
- DOPS: Learning to Detect 3D Objects and Predict Their 3D Shapes (CVPR 2020 · similarity 0.89)
- Towards Unsupervised Object Detection From LiDAR Point Clouds (CVPR 2023 · similarity 0.89)
- VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (CVPR 2018 · similarity 0.89)
- Voxel or Pillar: Exploring Efficient Point Cloud Representation for 3D Object Detection (AAAI 2024 · similarity 0.89)
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.
