Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.08917 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917374123311104 |
|---|---|
| author | Garg, Kapil Tang, Xinru Heo, Jimin Morgan, Dwayne R. Gergle, Darren Sudderth, Erik B. Piper, Anne Marie |
| author_facet | Garg, Kapil Tang, Xinru Heo, Jimin Morgan, Dwayne R. Gergle, Darren Sudderth, Erik B. Piper, Anne Marie |
| contents | Vision-Language Models (VLMs) are increasingly used by blind and low-vision (BLV) people to identify and understand products in their everyday lives, such as food, personal care items, and household goods. Despite their prevalence, we lack an empirical understanding of how common image quality issues--such as blur, misframing, and rotation--affect the accuracy of VLM-generated captions and whether the resulting captions meet BLV people's information needs. Based on a survey of 86 BLV participants, we develop an annotated dataset of 1,859 product images from BLV people to systematically evaluate how image quality issues affect VLM-generated captions. While the best VLM achieves 98% accuracy on images with no quality issues, accuracy drops to 75% overall when quality issues are present, worsening considerably as issues compound. We discuss the need for model evaluations that center on disabled people's experiences throughout the process and offer concrete recommendations for HCI and ML researchers to make VLMs more reliable for BLV people. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_08917 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | "It's trained by non-disabled people": Evaluating How Image Quality Affects Product Captioning with Vision-Language Models Garg, Kapil Tang, Xinru Heo, Jimin Morgan, Dwayne R. Gergle, Darren Sudderth, Erik B. Piper, Anne Marie Human-Computer Interaction Computer Vision and Pattern Recognition Vision-Language Models (VLMs) are increasingly used by blind and low-vision (BLV) people to identify and understand products in their everyday lives, such as food, personal care items, and household goods. Despite their prevalence, we lack an empirical understanding of how common image quality issues--such as blur, misframing, and rotation--affect the accuracy of VLM-generated captions and whether the resulting captions meet BLV people's information needs. Based on a survey of 86 BLV participants, we develop an annotated dataset of 1,859 product images from BLV people to systematically evaluate how image quality issues affect VLM-generated captions. While the best VLM achieves 98% accuracy on images with no quality issues, accuracy drops to 75% overall when quality issues are present, worsening considerably as issues compound. We discuss the need for model evaluations that center on disabled people's experiences throughout the process and offer concrete recommendations for HCI and ML researchers to make VLMs more reliable for BLV people. |
| title | "It's trained by non-disabled people": Evaluating How Image Quality Affects Product Captioning with Vision-Language Models |
| topic | Human-Computer Interaction Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.08917 |