Hugging Face's logo

Datasets:
j-min
/
layoutbench-coco

Tasks:
Text-to-Image
Languages:
English
ArXiv:
License:
Dataset card Data Studio Files
xet
Community
j-min commited on
Commit
7ad99d6
·
verified ·
1 Parent(s): 23e6b74

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -55,7 +55,7 @@ Authors:
55
  ## Summary
56
 
57
  LayoutBench-COCO is a diagnostic benchmark that examines layout-guided image generation models on arbitrary, unseen layouts.
58
- Instead of [LayoutBench](https://huggingface.co/datasets/j-min/layoutbench), LayoutBench-COCO consists of OOD layouts of real objects and suports zero-shot evaluation.
59
  LayoutBench-COCO measures 4 skills (Number, Position, Size, Combination), whose objects are from [MS COCO](https://cocodataset.org/).
60
  The new 'combination’ split consists of layouts with two objects in different spatial relations, and the remaining three splits are similar to those of LayoutBench.
61
 
 
55
  ## Summary
56
 
57
  LayoutBench-COCO is a diagnostic benchmark that examines layout-guided image generation models on arbitrary, unseen layouts.
58
+ Unlike [LayoutBench](https://huggingface.co/datasets/j-min/layoutbench), LayoutBench-COCO consists of OOD layouts of real objects and suports zero-shot evaluation.
59
  LayoutBench-COCO measures 4 skills (Number, Position, Size, Combination), whose objects are from [MS COCO](https://cocodataset.org/).
60
  The new 'combination’ split consists of layouts with two objects in different spatial relations, and the remaining three splits are similar to those of LayoutBench.
61