Research map / Detection and segmentation / Weakly supervised detection and segmentation
Change Is Everywhere: Single-Temporal Supervised Object Change Detection in Remote Sensing Imagery
Zhuo Zheng, Ailong Ma, Liangpei Zhang, Yanfei Zhong
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 semantic supervision to baseline (ablation) | ChangeStar baseline · LEVIR-CDall | IoU (%): 61.85 -> 62.42 | Improved |
| add semantic supervision to baseline with temporal symmetry (ablation) | ChangeStar · LEVIR-CDall | IoU (%): 64.10 -> 65.71 | Improved |
| add temporal symmetry to baseline (ablation) | ChangeStar baseline · LEVIR-CDall | IoU (%): 61.85 -> 64.10 | Improved |
| add temporal symmetry to baseline with semantic supervision (ablation) | ChangeStar · LEVIR-CDall | IoU (%): 62.42 -> 65.71 | Improved |
| use xor instead of or for label assignment (ablation) | ChangeStar · LEVIR-CDall | IoU (%): 43.84 -> 65.71 | Improved |
| ChangeMixin with over-deep conv subnetwork (N>=6) (ablation) | ChangeStar · LEVIR-CDall | IoU (%) | Hurt |
| increase ChangeMixin channels dc up to 80 and more (ablation) | ChangeStar · LEVIR-CDall | IoU (%) | Hurt |
| use ChangeStar instead of PCC | FarSeg · WHU (train xView2 pre-disaster) | IoU (%): 31.66 -> 58.22 | Improved |
| use ChangeStar instead of PCC | DeepLab v3+ · WHU (train xView2 pre-disaster) | IoU (%): 35.75 -> 52.01 | Improved |
| use ChangeStar instead of PCC | FarSeg · LEVIR-CDall (train SpaceNet2) | IoU (%): 7.97 -> 30.42 | Improved |
| train on xView2 pre-disaster instead of SpaceNet2 | ChangeStar (FarSeg) · LEVIR-CDall | IoU (%): 30.42 -> 65.71 | Improved |
| deeper backbone, ResNet-18 to ResNeXt-101 (single-temporal) | ChangeStar (FarSeg) · LEVIR-CDtest | IoU (%): 63.25 -> 68.84 | Improved |
Closest papers by meaning
- RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-𝑁 Ranking (NeurIPS 2026 · similarity 0.89)
- The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationf (CVPR 2025 · similarity 0.88)
- FM-ChangeNet: Learning Change through Pathwise Feature Transport (NeurIPS 2026 · similarity 0.88)
- Towards Generalizable Scene Change Detection (CVPR 2025 · similarity 0.88)
- Segment Any Change (NeurIPS 2024 · similarity 0.87)
- Feature Spectrum Learning for Remote Sensing Change Detection (CVPR 2025 · similarity 0.87)
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.
