Research map / Detection and segmentation / Open-set and few-shot detection
Single-Temporal Supervised Learning for Universal Remote Sensing Change Detection
Zhuo Zheng, Yanfei Zhong, Ailong Ma, Liangpei Zhang
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 single-temporal supervised learning with ChangeMixin (ChangeStar) over DPCC | FarSeg · LEVIR-CD all | F1: 71.0 -> 79.3 | Improved |
| add ChangeMixin2 and generalized STAR (ChangeStar2) over ChangeStar | FarSeg R-50 · S2Looking | F1: 21.2 -> 32.7 | Improved |
| add semantic supervision (ablation) | ChangeStar baseline · LEVIR-CD all | IoU: 61.9 -> 62.4 | Improved |
| add temporal symmetry (ablation) | ChangeStar baseline · LEVIR-CD all | F1: 76.4 -> 78.1 | Improved |
| add both semantic supervision and temporal symmetry (ablation) | ChangeStar · LEVIR-CD all | F1: 76.4 -> 79.3 | Improved |
| use xor instead of or label assignment (ablation) | ChangeStar · LEVIR-CD all | IoU: 43.8 -> 65.7 | Improved |
| add temporal difference network (ablation) | ChangeStar R-18 · LEVIR-CD test | F1: 77.4 -> 78.9 | Improved |
| add stochastic self-contrast (ablation) | ChangeStar R-18 + TDN · LEVIR-CD test | F1: 78.9 -> 81.2 | Improved |
| use stronger augmentation in stochastic self-contrast (ablation) | ChangeStar2 · LEVIR-CD test | F1: 81.2 -> 83.2 | Improved |
| reduce xView2 training data to 890 images | ChangeStar R-18 · LEVIR-CD test | F1 gap: -12.8 -> -20.6 | Hurt |
| use larger backbone with less data | ChangeStar RX-101 32x4d · LEVIR-CD test | F1 gap: -9.7 -> -11.5 | Mixed |
| single-temporal ChangeStar2 vs bitemporal supervised ChangeStar2 | FarSeg · DynamicEarthNet val* | SCS: 19.4 -> 20.2 | Improved |
Closest papers by meaning
- Change Is Everywhere: Single-Temporal Supervised Object Change Detection in Remote Sensing Imagery (ICCV 2021 · similarity 0.97)
- RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-𝑁 Ranking (NeurIPS 2026 · similarity 0.88)
- The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationf (CVPR 2025 · similarity 0.87)
- Feature Spectrum Learning for Remote Sensing Change Detection (CVPR 2025 · similarity 0.87)
- Towards Generalizable Scene Change Detection (CVPR 2025 · similarity 0.86)
- FM-ChangeNet: Learning Change through Pathwise Feature Transport (NeurIPS 2026 · 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.
