Research map / Detection and segmentation / 3D detection and driving
Adaptive LiDAR Scanning: Harnessing Temporal Cues for Efficient 3D Object Detection via Multi-Modal Fusion
Sara Shoouri, Morteza Tavakoli Taba, Hun-Seok Kim
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 adaptive LiDAR scanning (66.0% sparsity) | CMT R50 800x320 · nuScenes val next-frame | mAP (%): 67.9 -> 67.5 | Hurt |
| add adaptive LiDAR scanning (66.8% sparsity) | CMT VoVNet 1600x640 · nuScenes val next-frame | mAP (%): 70.3 -> 71.0 | Improved |
| add adaptive LiDAR scanning | CMT · Lyft val next-frame | w-mAP (%): 79.8 -> 80.4 | Improved |
| add adaptive LiDAR scanning | CMT* · nuScenes test | mAP (%): 70.4 -> 70.0 | Hurt |
| add adaptive scanning on all frames | CMT · Lyft val entire sequence | GFLOPs: 3351.2 -> 3289.6 | Improved |
| increase buffer length from 2 to 4 frames (ablation) | adaptive scan CMT · nuScenes val | mAP (%): 66.7 -> 67.5 | Improved |
| add distillation loss (ablation) | adaptive scan CMT · nuScenes val | mAP (%): 66.5 -> 67.3 | Improved |
| add CVaR loss (ablation) | adaptive scan CMT · nuScenes val | mAP (%): 66.5 -> 67.0 | Improved |
| add CVaR loss on top of distillation loss (ablation) | adaptive scan CMT · nuScenes val | mAP (%): 67.3 -> 67.5 | Improved |
| add differentiable voxelization (ablation) | adaptive scan CMT · nuScenes val | mAP (%): 66.7 -> 67.5 | Improved |
| add differentiable voxelization (ablation) | adaptive scan CMT · nuScenes val | scan sparsity (%): 68.1 -> 66.0 | Hurt |
| lower sparsity of adaptive scan (ablation) | adaptive scan CMT · nuScenes val | mAP and NDS | Improved |
Closest papers by meaning
- Detecting As Labeling: Rethinking LiDAR-camera Fusion in 3D Object Detection (ECCV 2024 · similarity 0.89)
- MOS: Model Synergy for Test-Time Adaptation on LiDAR-Based 3D Object Detection (ICLR 2025 · similarity 0.89)
- CCF: Complementary Collaborative Fusion for Domain Generalized Multi-Modal 3D Object Detection (CVPR 2026 · similarity 0.88)
- SparseLIF: High-Performance Sparse LiDAR-Camera Fusion for 3D Object Detection (ECCV 2024 · similarity 0.88)
- Approaching Outside: Scaling Unsupervised 3D Object Detection from 2D Scene (ECCV 2024 · similarity 0.88)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 2024 · similarity 0.87)
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.
