Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.00502 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916716023382016 |
|---|---|
| author | Ryu, Suho Kim, Kihyun Baek, Eugene Shin, Dongsoo Lee, Joonseok |
| author_facet | Ryu, Suho Kim, Kihyun Baek, Eugene Shin, Dongsoo Lee, Joonseok |
| contents | A variety of text-guided image editing models have been proposed recently. However, there is no widely-accepted standard evaluation method mainly due to the subjective nature of the task, letting researchers rely on manual user study. To address this, we introduce a novel Human-Aligned benchmark for Text-guided Image Editing (HATIE). Providing a large-scale benchmark set covering a wide range of editing tasks, it allows reliable evaluation, not limited to specific easy-to-evaluate cases. Also, HATIE provides a fully-automated and omnidirectional evaluation pipeline. Particularly, we combine multiple scores measuring various aspects of editing so as to align with human perception. We empirically verify that the evaluation of HATIE is indeed human-aligned in various aspects, and provide benchmark results on several state-of-the-art models to provide deeper insights on their performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_00502 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Towards Scalable Human-aligned Benchmark for Text-guided Image Editing Ryu, Suho Kim, Kihyun Baek, Eugene Shin, Dongsoo Lee, Joonseok Computer Vision and Pattern Recognition A variety of text-guided image editing models have been proposed recently. However, there is no widely-accepted standard evaluation method mainly due to the subjective nature of the task, letting researchers rely on manual user study. To address this, we introduce a novel Human-Aligned benchmark for Text-guided Image Editing (HATIE). Providing a large-scale benchmark set covering a wide range of editing tasks, it allows reliable evaluation, not limited to specific easy-to-evaluate cases. Also, HATIE provides a fully-automated and omnidirectional evaluation pipeline. Particularly, we combine multiple scores measuring various aspects of editing so as to align with human perception. We empirically verify that the evaluation of HATIE is indeed human-aligned in various aspects, and provide benchmark results on several state-of-the-art models to provide deeper insights on their performance. |
| title | Towards Scalable Human-aligned Benchmark for Text-guided Image Editing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.00502 |