Research map / Label-efficient and robust learning / Domain adaptation and generalization
Dynamic to Static Lidar Scan Reconstruction Using Adversarially Trained Auto Encoder
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 |
|---|---|---|---|
| Use adversarial training with a pair discriminator to map dynamic scans to static scans | DSLR · CARLA-64 | Chamfer: 4.05 -> 1.00 | Helped |
| Use adversarial training with a pair discriminator to map dynamic scans to static scans | DSLR · ARD-16 | Chamfer: 0.38 -> 0.20 | Helped |
| Add segmentation-based reconstruction to DSLR (ablation) | DSLR · CARLA-64 | Chamfer: 1.00 -> 0.02 | Helped |
| Train DSLR on the improved dataset created with segmentation-based dynamic-point replacement (ablation) | DSLR · CARLA-64 | Chamfer: 1.00 -> 0.49 | Helped |
| Add unsupervised domain adaptation (ablation) | DSLR · KITTI-64 | LQI: 1.120 -> 1.119 | Helped |
| Use DSLR-Seg with ground-truth segmentation masks (ablation) | DSLR-Seg · CARLA-64 SLAM | ATE: 10.360 -> 7.330 | Helped |
| Use DSLR-Seg with U-Net-predicted segmentation masks (ablation) | DSLR-Seg · KITTI-64 SLAM | ATE: 11.809 -> 99.642 | Hurt |
| Reconstruct scans with DSLR rather than using the unprocessed dynamic scan | DSLR · ARD-16 SLAM | ATE: 1.701 -> 1.680 | Helped |
| Reconstruct scans with DSLR rather than using the unprocessed dynamic scan | DSLR · ARD-16 SLAM | Rot RPE: 0.613\textdegree -> 0.614\textdegree | Hurt |
Closest papers by meaning
- 3D LiDAR Mapping in Dynamic Environments using a 4D Implicit Neural Representation (CVPR 2024 · similarity 0.88)
- Dynamic LiDAR Re-simulation using Compositional Neural Fields (CVPR 2024 · similarity 0.86)
- LiT: Unifying LiDAR "Languages" with LiDAR Translator (NeurIPS 2024 · similarity 0.85)
- Swift-Mapping: Online Neural Implicit Dense Mapping in Urban Scenes (AAAI 2024 · similarity 0.84)
- ICP-Flow: LiDAR Scene Flow Estimation with ICP (CVPR 2024 · similarity 0.84)
- D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation (CVPR 2025 · similarity 0.84)
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.