You're currently viewing an old version of this dataset. To see the current version, click here.

T2I-Compbench-Count

A benchmark for open-world compositional text-to-image generation. The dataset consists of 218 prompts with a single object and its number.

Data and Resources

Cite this as

Kaiyi Huang, Kaiyue Sun, Enze Xie, Zhenguo Li, Xihui Liu (2024). Dataset: T2I-Compbench-Count. https://doi.org/10.57702/2n318phy

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2406.10210
Author Kaiyi Huang
More Authors
Kaiyue Sun
Enze Xie
Zhenguo Li
Xihui Liu
Homepage https://make-it-count-paper.github.io