Research map / Point clouds / Segmentation and detection
Associate-3Ddet: Perceptual-to-Conceptual Association for 3D Point Cloud Object Detection
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 |
|---|---|---|---|
| associate perceptual features with features from complete, class-wise conceptual models using P2C and CFG (ablation) | voxel-based 3D detector · KITTI val split, Car | 3D AP moderate: 76.78 -> 79.17 | Helped |
| add deformable convolution without P2C and CFG (ablation) | Baseline · KITTI val split, Car | 3D AP moderate: 76.78 -> 76.86 | No clear effect |
| add the P2C and CFG association mechanism (ablation) | Baseline · KITTI val split, Car | 3D AP moderate: 76.78 -> 78.05 | Helped |
| add deformable convolution to P2C and CFG (ablation) | Associate-3Ddet · KITTI val split, Car | 3D AP moderate: 78.05 -> 78.29 | Helped |
| add the foreground mask to P2C and CFG (ablation) | Associate-3Ddet · KITTI val split, Car | 3D AP moderate: 78.05 -> 78.46 | Helped |
| add the foreground mask and deformable convolution (ablation) | Associate-3Ddet · KITTI val split, Car | 3D AP moderate: 78.29 -> 78.69 | Helped |
| add incompletion-aware offset reweighting to the foreground-masked adaptation (ablation) | Associate-3Ddet · KITTI val split, Car | 3D AP moderate: 78.69 -> 79.17 | Helped |
| mix real and conceptual data without siamese domain adaptation (ablation) | Baseline · KITTI val split, Car | 3D AP moderate: 76.78 -> 76.75 | No clear effect |
| select the top 20% rather than top 50% of instances as conceptual models (ablation) | Associate-3Ddet conceptual-scene construction · KITTI val split, Car | 3D AP moderate: 78.45 -> 79.17 | Helped |
| select only the top 10% rather than top 20% of instances as conceptual models (ablation) | Associate-3Ddet conceptual-scene construction · KITTI val split, Car | 3D AP moderate: 79.17 -> 78.89 | Hurt |
Closest papers by meaning
- One for All: Multi-Domain Joint Training for Point Cloud Based 3D Object Detection (NeurIPS 2024 · similarity 0.89)
- Towards Learning Group-Equivariant Features for Domain Adaptive 3D Detection (NeurIPS 2024 · similarity 0.88)
- Unified Domain Generalization and Adaptation for Multi-View 3D Object Detection (NeurIPS 2024 · similarity 0.87)
- Pseudo Label Refinery for Unsupervised Domain Adaptation on Cross-dataset 3D Object Detection (CVPR 2024 · similarity 0.87)
- CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-Based 3D Object Detection (AAAI 2024 · similarity 0.86)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 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.