Research map / Detection and segmentation / Weakly supervised detection and segmentation
Fully Convolutional Change Detection Framework With Generative Adversarial Network for Unsupervised, Weakly Supervised and Regional Supervised Change Detection
Chen Wu, Bo Du, 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 |
|---|---|---|---|
| replace RNN/SiamCRNN pre-detection pipeline with end-to-end FCD-GAN | unsupervised change detection · WH | F1: 0.7383 -> 0.7756 | Improved |
| replace SiamCRNN_LSTM with FCD-GAN | unsupervised change detection · HY | F1: 0.7957 -> 0.8409 | Improved |
| add generation loss to adversarial-only model (threshold 0.9) (ablation) | FCD-AN · BCD | F1: 0.5468 -> 0.5645 | Improved |
| add generation loss to adversarial-only model (threshold 0.5) (ablation) | FCD-AN · BCD | F1: 0.5272 -> 0.5452 | Improved |
| add discriminator/adversarial loss to generation-only model (threshold 0.9) (ablation) | FCD-GN · BCD | F1: 0.4407 -> 0.5645 | Improved |
| add discriminator to generation-only model (threshold 0.5) (ablation) | FCD-GN · BCD | F1: 0.4141 -> 0.5452 | Improved |
| raise evaluation threshold from 0.5 to 0.9 | FCD-GAN · BCD | F1: 0.5452 -> 0.5645 | Improved |
| replace CAM with FCD-GAN (threshold 0.9) | weakly supervised segmentation · BCD | F1: 0.5460 -> 0.5645 | Improved |
| add generator to discriminator-only model (threshold 0.3) (ablation) | FCD-AN · OSCD | F1: 0.3055 -> 0.3508 | Improved |
| add discriminator to generation-only model (threshold 0.3) (ablation) | FCD-GN · OSCD | F1: 0.3173 -> 0.3508 | Improved |
| lower threshold from 0.5 to 0.3 | FCD-GAN · OSCD | F1: 0.3369 -> 0.3508 | Improved |
| use full pixel-level supervision (FSCD) instead of region-level supervision | segmentor · OSCD | F1: 0.4044 vs 0.3508 | Hurt |
Closest papers by meaning
- Feature Spectrum Learning for Remote Sensing Change Detection (CVPR 2025 · similarity 0.88)
- The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationf (CVPR 2025 · similarity 0.85)
- MUCD: Unsupervised Point Cloud Change Detection via Masked Consistency (AAAI 2025 · similarity 0.85)
- Towards Generalizable Scene Change Detection (CVPR 2025 · similarity 0.84)
- FM-ChangeNet: Learning Change through Pathwise Feature Transport (NeurIPS 2026 · similarity 0.84)
- ChangeDiff: A Multi-Temporal Change Detection Data Generator with Flexible Text Prompts via Diffusion Model (AAAI 2025 · 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.
