Skip to content

The Database of "Hierarchical Scene Parsing by Weakly Supervised Learning with Image Descriptions", IEEE TPAMI 2018

Notifications You must be signed in to change notification settings

ruixuejianfei/SYSU-Scene

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SYSU-Scene Dataset

SYSU-Scene example

SYSU-Scene dataset is a new parsing database including elaborately annotated objects.

  • 5,046 high-resolution photos with totally 37 tags
  • Wide range of categories, relationship, and scenes
  • All images are with pixel-level annotations

Feel free to contact [email protected] should you have any suggestions or questions.

Please visit the project page for more information.

Files

Root directory contains following files and folders:

  • ground truth/ - directory of annotations (5046 files.mat)
  • gtvis/ - directory of visual result of annotations (5046 files.png)
  • images/ - directory of original photos (5046 files.jpg)
  • seg/ - directory of segmentations (5046 files.png)
  • label_list.mat - [1*37] cell array which maps label numbers to label names while '0' maps to 'background'
  • README.md - this file
  • sample.png - sample annotations
  • sentences.json - semantic description of images

Notes on Image-Level Annotations

Each annotation is saved in ground truth/ as a matlab .mat file, which is a variable

  • groundtruth: [h*w] matrix denotes the annotated labels of pixels

About

The Database of "Hierarchical Scene Parsing by Weakly Supervised Learning with Image Descriptions", IEEE TPAMI 2018

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages