Research map / Point clouds / Segmentation and detection
Learning to Suppress SPAD-based LiDAR Flare
Xuanya Zhu, Linghao Shen
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 second echo and ambient illumination to the input (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 75.18 -> 78.49 | Improved |
| add ambient illumination x0 on top of second echo (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 75.56 -> 78.49 | Improved |
| add second echo x2 on top of ambient illumination (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 77.67 -> 78.49 | Improved |
| use multi-echo 7-channel input instead of first-echo input | DeepLabV3+ · FLARE dataset (SPAD LiDAR flare segmentation) | mIoU: 69.28 -> 72.14 | Improved |
| use multi-echo 7-channel input instead of first-echo input | OCRNet · FLARE dataset (SPAD LiDAR flare segmentation) | mIoU: 68.37 -> 69.59 | Improved |
| add FlareAug crop operation (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 75.58 -> 78.49 | Improved |
| add FlareAug fuse operation (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 77.41 -> 78.49 | Improved |
| add both crop and fuse augmentation (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 74.91 -> 78.49 | Improved |
| add Physics-Aware Encoder (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 64.41 -> 71.01 | Improved |
| add Echo-Aware Fusion (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 71.01 -> 75.40 | Improved |
| add attention decoder (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 75.40 -> 78.49 | Improved |
| set depth bins N=16 instead of 4, 8 or 32 (ablation) | PILF · FLARE dataset (SPAD LiDAR flare segmentation) (ablation, 5 runs) | mIoU: 78.21, 77.87, 77.28 -> 78.49 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- RandLANet — RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds (CVPR 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Light Source Guided Single-Image Flare Removal From Unpaired Data (ICCV 2021 · similarity 0.86)
- Disentangle Nighttime Lens Flares: Self-supervised Generation-based Lens Flare Removal (AAAI 2025 · similarity 0.86)
- FlareX: A Physics-Informed Dataset for Lens Flare Removal via 2D Synthesis and 3D Rendering (NeurIPS 2025 · similarity 0.85)
- Language-Guided One-Step Diffusion Model for Nighttime Flare Removal (CVPR 2026 · similarity 0.84)
- FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Views (CVPR 2025 · similarity 0.84)
- LiT: Unifying LiDAR "Languages" with LiDAR Translator (NeurIPS 2024 · similarity 0.83)
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.
