Change Detection
Change Detection is a computer vision task that involves detecting changes in an image or video sequence over time. The goal is to identify areas in the image or video that have undergone changes, such as appearance changes, object disappearance or appearance, or even changes in the scene's background.
Image credit: "A TRANSFORMER-BASED SIAMESE NETWORK FOR CHANGE DETECTION"
Papers
Showing 1–10 of 919 papers
All datasetsLEVIR-CDWHU-CDCDD Dataset (season-varying)SYSU-CDS2LookingDSIFN-CDLEVIR+WHU Building DatasetOSCD - 13chOSCD - 3chGoogleGZ-CDGVLM
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | SChanger-base | F1-Score | 68.95 | — | Unverified |
| 2 | SChanger-small | F1-Score | 68.2 | — | Unverified |
| 3 | ChangeStar(1x96, MiT-b1, Changen-90k) | F1-Score | 67.9 | — | Unverified |
| 4 | LSKNet-S | F1-Score | 67.52 | — | Unverified |
| 5 | ChangeStar(1x96, ResNet-18, Changen-90k) | F1-Score | 67.1 | — | Unverified |
| 6 | Changer (Ex, ResNet-18) | F1-Score | 66.2 | — | Unverified |
| 7 | CGNet | F1-Score | 64.33 | — | Unverified |
| 8 | BD-MSA | F1-Score | 64.08 | — | Unverified |
| 9 | HCGMNet | F1-Score | 63.87 | — | Unverified |
| 10 | C2FNet | F1-Score | 62.83 | — | Unverified |