Papers › HCGMNET: A Hierarchical Change Guiding Map Network For Change Detection

HCGMNET: A Hierarchical Change Guiding Map Network For Change Detection

21 Feb 2023arXiv:2302.10420archive 2025-07-28

Chengxi Han, Chen Wu, Bo Du

Very-high-resolution (VHR) remote sensing (RS) image change detection (CD) has been a challenging task for its very rich spatial information and sample imbalance problem. In this paper, we have proposed a hierarchical change guiding map network (HCGMNet) for change detection. The model uses hierarchical convolution operations to extract multiscale features, continuously merges multi-scale features layer by layer to improve the expression of global and local information, and guides the model to gradually refine edge features and comprehensive performance by a change guide module (CGM), which is a self-attention with changing guide map. Extensive experiments on two CD datasets show that the proposed HCGMNet architecture achieves better CD performance than existing state-of-the-art (SOTA) CD methods.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

ChengxiHAN/HCGMNet-CD officialpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Change Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Change Detection CDD Dataset (season-varying) HCGMNet F1 95.07 #14 of 18 Archive leaderboard report
Change Detection CDD Dataset (season-varying) HCGMNet F1-Score 95.07 #14 of 18 Archive leaderboard report
Change Detection CDD Dataset (season-varying) HCGMNet IoU 90.60 #14 of 18 Archive leaderboard report
Change Detection CDD Dataset (season-varying) HCGMNet KC 94.40 #14 of 18 Archive leaderboard report
Change Detection CDD Dataset (season-varying) HCGMNet Overall Accuracy 98.82 #14 of 18 Archive leaderboard report
Change Detection CDD Dataset (season-varying) HCGMNet Precision 93.84 #14 of 18 Archive leaderboard report
Change Detection CDD Dataset (season-varying) HCGMNet Recall 96.34 #14 of 18 Archive leaderboard report
Change Detection DSIFN-CD HCGMNet F1 55.00 #9 of 9 Archive leaderboard report
Change Detection DSIFN-CD HCGMNet IoU 37.93 #9 of 9 Archive leaderboard report
Change Detection DSIFN-CD HCGMNet KC 41.53 #9 of 9 Archive leaderboard report
Change Detection DSIFN-CD HCGMNet Overall Accuracy 76.26 #9 of 9 Archive leaderboard report
Change Detection DSIFN-CD HCGMNet Precision 40.57 #9 of 9 Archive leaderboard report
Change Detection DSIFN-CD HCGMNet Recall 85.35 #9 of 9 Archive leaderboard report
Change Detection GoogleGZ-CD HCGMNet F1 85.71 #3 of 4 Archive leaderboard report
Change Detection GoogleGZ-CD HCGMNet IoU 74.99 #3 of 4 Archive leaderboard report
Change Detection GoogleGZ-CD HCGMNet KC 80.94 #3 of 4 Archive leaderboard report
Change Detection GoogleGZ-CD HCGMNet Overal Accuracy 92.85 #3 of 4 Archive leaderboard report
Change Detection GoogleGZ-CD HCGMNet Precision 84.25 #3 of 4 Archive leaderboard report
Change Detection GoogleGZ-CD HCGMNet Recall 87.22 #3 of 4 Archive leaderboard report
Change Detection LEVIR+ HCGMNet F1 82.37 #7 of 9 Archive leaderboard report
Change Detection LEVIR+ HCGMNet IoU 70.03 #7 of 9 Archive leaderboard report
Change Detection LEVIR+ HCGMNet KC 81.63 #7 of 9 Archive leaderboard report
Change Detection LEVIR+ HCGMNet OA 98.57 #7 of 9 Archive leaderboard report
Change Detection LEVIR+ HCGMNet Prcision 82.81 #7 of 9 Archive leaderboard report
Change Detection LEVIR+ HCGMNet Recall 81.94 #7 of 9 Archive leaderboard report
Change Detection LEVIR-CD HCGMNet F1 91.77 #17 of 28 Archive leaderboard report
Change Detection LEVIR-CD HCGMNet F1-score 91.77 #17 of 28 Archive leaderboard report
Change Detection LEVIR-CD HCGMNet IoU 84.79 #17 of 28 Archive leaderboard report
Change Detection LEVIR-CD HCGMNet Overall Accuracy 99.18 #17 of 28 Archive leaderboard report
Change Detection LEVIR-CD HCGMNet Precision 92.96 #17 of 28 Archive leaderboard report
Change Detection LEVIR-CD HCGMNet Recall 90.61 #17 of 28 Archive leaderboard report
Change Detection S2Looking HCGMNet F1 63.87 #9 of 11 Archive leaderboard report
Change Detection S2Looking HCGMNet F1-Score 63.87 #9 of 11 Archive leaderboard report
Change Detection S2Looking HCGMNet IoU 46.91 #9 of 11 Archive leaderboard report
Change Detection S2Looking HCGMNet KC 63.48 #9 of 11 Archive leaderboard report
Change Detection S2Looking HCGMNet OA 99.22 #9 of 11 Archive leaderboard report
Change Detection S2Looking HCGMNet Precision 72.51 #9 of 11 Archive leaderboard report
Change Detection S2Looking HCGMNet Recall 57.06 #9 of 11 Archive leaderboard report
Change Detection SYSU-CD HCGMNet F1 79.76 #9 of 12 Archive leaderboard report
Change Detection SYSU-CD HCGMNet IoU 66.33 #9 of 12 Archive leaderboard report
Change Detection SYSU-CD HCGMNet KC 74.11 #9 of 12 Archive leaderboard report
Change Detection SYSU-CD HCGMNet OA 91.12 #9 of 12 Archive leaderboard report
Change Detection SYSU-CD HCGMNet Precision 86.28 #9 of 12 Archive leaderboard report
Change Detection SYSU-CD HCGMNet Recall 74.15 #9 of 12 Archive leaderboard report
Change Detection WHU-CD HCGMNet F1 92.08 #12 of 22 Archive leaderboard report
Change Detection WHU-CD HCGMNet IoU 85.33 #12 of 22 Archive leaderboard report
Change Detection WHU-CD HCGMNet KC 91.80 #12 of 22 Archive leaderboard report
Change Detection WHU-CD HCGMNet Overall Accuracy 99.45 #12 of 22 Archive leaderboard report
Change Detection WHU-CD HCGMNet Precision 93.93 #12 of 22 Archive leaderboard report
Change Detection WHU-CD HCGMNet Recall 90.31 #12 of 22 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Convolution

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections