Skip to content

Latest commit

 

History

History
71 lines (67 loc) · 2.98 KB

README.md

File metadata and controls

71 lines (67 loc) · 2.98 KB
Raw

Prepare the datasets

✈️ 1: Prepare the binary change detection datasets:

  • Download datasets LEVIR, LEVIR+, S2Looking, and SYSU
  • Crop LEVIR dataset into 256x256 patches.
  • Generate list file as ls -R ./label/* > test.txt
  • Prepare datasets into the following structure and set their path in ./lib/configs/base/datasets/levir_congfig.py, ./lib/configs/base/datasets/levir_plus_congfig.py, ./lib/configs/base/datasets/s2looking_congfig.py, and ./lib/configs/base/datasets/sysu_congfig.py 
    ├─train
    ├─A        ...jpg/png
    ├─B        ...jpg/png
    ├─label    ...jpg/png
    └─list     ...txt
├─val
    ├─A
    ├─B
    ├─label
    └─list
├─test
    ├─A
    ├─B
    ├─label
    └─list

✈️ 2: Prepare the semantic change detection datasets:

  • Download datasets SECOND, and Landsat-SCD
  • The pre-processed datasets can be obtained from SECOND, and the train and test list can be downloaded from list.
  • The pre-processed Landsat-SCD dataset can be obtained from Landsat-SCD.
  • Prepare datasets into the following structure and set their path in ./lib/configs/base/datasets/second_congfig.py and ./lib/configs/base/datasets/landsatcd_congfig.py
├─train
    ├─im1        ...jpg/png
    ├─im2        ...jpg/png
    ├─label1    ...jpg/png
    ├─label2    ...jpg/png
    └─list     ...txt
├─val
    ├─im1        
    ├─im2        
    ├─label1    
    ├─label2   
    └─list  
├─test
    ├─im1        
    ├─im2        
    ├─label1    
    ├─label2   
    └─list

✈️ 3:Prepare the building damage assessment dataset:

  • Download dataset xView2.
  • Create the rbg mask from xView2 by using './data/xBD/preoricess_data.py'.
  • Create the image list from xView2 by using './data/xBD/capture_image_list.py'.
  • Prepare the dataset into the following structure and set its path in ./lib/configs/base/datasets/xview2_congfig.py
├─train
    ├─images        ...jpg/png
    ├─labels        ...json
    ├─masks    ...jpg/png
    ├─targets    ...jpg/png
    ├─image_list.txt     
├─test
    ├─images        ...jpg/png
    ├─labels        ...json
    ├─masks    ...jpg/png
    ├─targets    ...jpg/png
    ├─image_list.txt