Research map / Detection and segmentation / CNN features
DSA: Discriminative Scatter Analysis for Early Smoke Segmentation
Lujian Yao, Haitao Zhao, Jingchao Peng, Zhongze Wang, Kaijie Zhao
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 |
|---|---|---|---|
| use scatter-based IOS instead of distance-based loss (ablation) | SegFormer MiT-B3 · SmokeSeg | F1: 68.25 -> 69.24 | Improved |
| add OOS (proxy-proxy scatter) alone (ablation) | SegFormer MiT-B3 · SmokeSeg | F1: 67.99 -> 68.61 | Improved |
| combine IOS and OOS (ablation) | SegFormer MiT-B3 · SmokeSeg | F1: 69.24 -> 70.15 | Improved |
| add DSA to existing segmentation models | FoSp MiT-B3 · SmokeSeg | F1: 72.05 -> 73.40 | Improved |
| add DSA to real-time segmentation models | AFFormer and SegFormer MiT-B0 · SmokeSeg | F1: 60.17 -> 61.82; 60.73 -> 62.23 | Improved |
| add ProxyAnchor loss to FoSp (ablation) | FoSp MiT-B3 · SmokeSeg | F1: 72.05 -> 70.30 | Hurt |
| add triplet loss to FoSp (ablation) | FoSp MiT-B3 · SmokeSeg | F1: 72.05 -> 71.89 | Hurt |
| add SoftTriple multi-proxy loss (ablation) | FoSp MiT-B3 · SmokeSeg | F1: 72.05 -> 72.50 | Improved |
| increase proxies per class from 1 to 3 and then 5 (ablation) | DSA · SmokeSeg | F1 | Improved |
| add DSA to general smoke segmentation on SMOKE5K | SegFormer MiT-B3 · SMOKE5K | F1: 78.6 -> 79.5 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- FoSp — FoSp: Focus and Separation Network for Early Smoke Segmentation (AAAI 2024) · better on 5 of 5 shared results
- SegFormer — SegFormer: A Topic Segmentation Model with Controllable Range of Attention (AAAI 2023) · better on 3 of 3 shared results
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
- FoSp: Focus and Separation Network for Early Smoke Segmentation (AAAI 2024 · similarity 0.89)
- CoSW: Conditional Sample Weighting for Smoke Segmentation with Label Noise (NeurIPS 2024 · similarity 0.86)
- DetectiumFire: A Comprehensive Multi-modal Dataset Bridging Vision and Language for Fire Understanding (NeurIPS 2025 · similarity 0.81)
- FireScope: Wildfire Risk Raster Prediction With a Chain-of-Thought Oracle (CVPR 2026 · similarity 0.80)
- Guided Real Image Dehazing Using YCbCr Color Space (AAAI 2025 · similarity 0.80)
- ROS-SAM: High-Quality Interactive Segmentation for Remote Sensing Moving Object (CVPR 2025 · similarity 0.80)
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.
