Skip to content

Latest commit

 

History

History
48 lines (44 loc) · 2.7 KB

data.md

File metadata and controls

48 lines (44 loc) · 2.7 KB
Raw

Data Preparation

The datasets we used for five different SOD tasks are as follows:

Task         Train sets Test sets
RGB [cr] DUTS-TR [ce] HKU-IS, PASCAL-S, ECSSD, DUTS-TE, DUT-OMRON, MSB-TE
RGB-D [dr] RGBD-TR [de] DUT, LFSD, NJUD, NLPR, RGBD135, SIP, SSD, STERE1000, STEREO
RGB-T [tr] VT5000-TR [te] VT821, VT1000 and VT5000-TE
Video [or] VSOD-TR [oe] SegV2, FBMS, DAVIS-TE, DAVSOD-TE
RSI(gvoh) [rr] RSSD-TR [re] ORSSD, EORSSD, ORS

RGBD-TR (2985 samples) contains 1,485 images from NJUD, 700 images from NLPR, and 800 images from DUTLF-Depth.
VT5000-TR and VT5000-TE are the train and test splits of the VT5000 dataset.
VSOD-TR is the collection of the train splits of the DAVIS and DAVSOD datasets.
RSSD-TR (4000 samples) contains 2000 images from ORS, 1400 images from EORSSD, and 600 images from ORSSD.

The dataset we used is consistent with the existing mainstream methods of five SOD tasks.

Employment and Customization

Your /datasets folder should look like this:

-- datasets
   |-- DUT-O
   |   |--RGB
   |   |--GT
   |-- DUTS-TR
   |   |--RGB
   |   |--GT
   |-- NJUD-TE
   |   |--RGB
   |   |--GT
   ...

For video and RSI data, we nested an additional folder, which appears as:

-- datasets
   |-- VSOD
   |   |--FBMS
   |   |--SegV2
   |   ...
   |-- ORSSD
   |   |--ORSSD-TE
   |   |--RSSD-TR
   |   ...
   ...

You can also modify the get_train_image_list(), get_test_list(), get_rgbd_list() and other method in data.py to specify the dataset and path to be used. This work involves multiple tasks and datasets, so feel free to adjust according to your specific needs.

If you wish to transfer the model to a new SOD task, you can follow the logic of the get_remote_list() method to write a new method for obtaining the dataset list for that specific task, choosing a character as the identifier for that task. Finally, incorporate the logic for the corresponding task in get_train_image_list() and get_test_image_list().