Research map / Detection and segmentation / 3D detection and driving
GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving
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 visible-surface features to guidance-box features (ablation) | GS3D refinement · KITTI val_2, car | AP3D at IoU=0.5, Moderate: 15.47 -> 20.41 | Helped |
| Replace residual regression with classification refinement while using surface features (ablation) | GS3D refinement · KITTI val_2, car | AP3D at IoU=0.5, Moderate: 20.41 -> 23.39 | Helped |
| Replace fixed-feature classification with shift-feature classification (ablation) | GS3D refinement · KITTI val_2, car | AP3D at IoU=0.7, Moderate: 3.76 -> 4.51 | Helped |
| Replace fixed-feature classification with shift-feature classification (ablation) | GS3D refinement · KITTI val_2, car | AP3D at IoU=0.5, Moderate: 23.39 -> 18.81 | Hurt |
| Add quality-aware loss to classification refinement (ablation) | GS3D with surface features and fixed-feature classification · KITTI val_2, car | AP3D at IoU=0.5, Moderate: 23.39 -> 27.16 | Helped |
| Add quality-aware loss to shift-feature classification refinement (ablation) | GS3D with surface features and shift-feature classification · KITTI val_2, car | AP3D at IoU=0.7, Moderate: 4.51 -> 10.51 | Helped |
| Use shift-based rather than fixed-feature classification (ablation) | GS3D on KITTI validation sets · KITTI val_1 and val_2, car | AP3D at IoU=0.7: 8.71 -> 11.63; 6.64 -> 10.51 | Helped |
| Use shift-based rather than fixed-feature classification (ablation) | GS3D on KITTI validation sets · KITTI val_1 and val_2, car | AP3D at IoU=0.5: 33.11 -> 30.60; 27.16 -> 26.40 | Mixed |
| Use classification refinement with quality-aware loss instead of direct regression (ablation) | GS3D on KITTI val_2 · KITTI val_2, car | AP3D at IoU=0.7, Moderate: 1.99 -> 6.64 | Helped |
| Use shift-feature classification with quality-aware loss instead of direct regression (ablation) | GS3D on KITTI val_2 · KITTI val_2, car | AP3D at IoU=0.7, Moderate: 1.99 -> 10.51 | Helped |
Closest papers by meaning
- OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection (NeurIPS 2025 · similarity 0.87)
- Cubify Anything: Scaling Indoor 3D Object Detection (CVPR 2025 · similarity 0.87)
- OpenBox: Annotate Any Bounding Boxes in 3D (NeurIPS 2025 · similarity 0.87)
- DriveGEN: Generalized and Robust 3D Detection in Driving via Controllable Text-to-Image Diffusion Generation (CVPR 2025 · similarity 0.87)
- Training an Open-Vocabulary Monocular 3D Detection Model without 3D Data (NeurIPS 2024 · similarity 0.86)
- Geometry-Guided Domain Generalization for Monocular 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.