README.md

metadata

license: cc-by-nc-sa-4.0

Dataset for Paper: Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Vision.

Images: images.tar

dev-labels: llvisionqa_dev.json

test-labels: llvisionqa_test.json

See Github for Usage: https://github.com/vqassessment/q-bench.

Feel free to cite us.

@article{wu2023qbench,
    title={Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Vision},
    author={Wu, Haoning and Zhang, Zicheng and Zhang, Erli and Chen, Chaofeng and Liao, Liang and Wang, Annan and Li, Chunyi and Sun, Wenxiu and Yan, Qiong and Zhai, Guangtao and Lin, Weisi},
    year={2023},
    eprint={2309.14181},
}