Research map / Detection and segmentation / 3D detection and driving
AFDetV2: Rethinking the Necessity of the Second Stage for Object Detection from Point Clouds
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 keypoint auxiliary loss (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 64.65 -> 64.77 | Helped |
| Add SC-Conv backbone (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 64.65 -> 65.20 | Helped |
| Add IoU prediction and rescoring (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 64.65 -> 67.77 | Helped |
| Add keypoint auxiliary loss, SC-Conv backbone, and IoU rescoring (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 64.65 -> 68.77 | Helped |
| Add IoU-aware rescoring (ablation) | AFDetV2-PP · Waymo Open Dataset val | ALL APH/L2: 63.15 -> 64.57 | Helped |
| Add second-stage box refinement (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 68.81 -> 68.79 | No clear effect |
| Add second-stage score refinement together with box refinement (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 68.81 -> 70.42 | Helped |
| Use single-stage IoU-aware rescoring instead of second-stage box and score refinement (ablation) | AFDetV2 · Waymo Open Dataset val | ALL APH/L2: 70.42 -> 71.10 | Helped |
| Add SC-Conv backbone (ablation) | Voxel R-CNN · Waymo Open Dataset val subset | ALL APH/L2: 53.23 -> 57.14 | Helped |
| Add second-stage box refinement (ablation) | Voxel R-CNN with SC-Conv · Waymo Open Dataset val subset | ALL APH/L2: 57.14 -> 57.69 | No clear effect |
| Add SC-Conv and IoU rescoring and remove second stage (ablation) | Voxel R-CNN · Waymo Open Dataset val subset | ALL APH/L2: 53.23 -> 59.88 | Helped |
| Add SC-Conv and IoU branch while retaining second-stage refinements (ablation) | Voxel R-CNN · Waymo Open Dataset val subset | ALL APH/L2: 58.78 -> 59.56 | Helped |
Closest papers by meaning
- ImOV3D: Learning Open Vocabulary Point Clouds 3D Object Detection from Only 2D Images (NeurIPS 2024 · similarity 0.87)
- LION: Linear Group RNN for 3D Object Detection in Point Clouds (NeurIPS 2024 · similarity 0.87)
- Point2Real: Bridging the Gap between Point Cloud and Realistic Image for Open-World 3D Recognition (AAAI 2024 · similarity 0.86)
- SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection (CVPR 2024 · similarity 0.86)
- One for All: Multi-Domain Joint Training for Point Cloud Based 3D Object Detection (NeurIPS 2024 · similarity 0.86)
- AS-Det: Active Sampling for Adaptive 3D Object Detection in Point Clouds (AAAI 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.