Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909001986342912 |
|---|---|
| author | Yu, Hong-Tao Peng, Yuxin Belongie, Serge Wei, Xiu-Shen |
| author_facet | Yu, Hong-Tao Peng, Yuxin Belongie, Serge Wei, Xiu-Shen |
| contents | Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal perception capabilities, garnering significant attention. While numerous evaluation studies have emerged, assessing LVLMs both holistically and on specialized tasks, fine-grained image tasks-fundamental to computer vision-remain largely unexplored. To fill this gap, we introduce a comprehensive fine-grained evaluation benchmark, i.e., FG-BMK, comprising 1.01 million questions and 0.33 million images. Our evaluation systematically examines LVLMs from both human-oriented and machine-oriented perspectives, focusing on their semantic recognition and fine-grained feature representation capabilities. Through extensive experiments on twelve representative LVLMs/VLMs, we uncover key findings regarding the influence of training paradigms, modality alignment, perturbation susceptibility, and fine-grained category reasoning on task performance. This work provides critical insights into the limitations of current LVLMs and offers guidance for future data construction and model design in the development of more advanced LVLMs. Our code is open-source and available at https://github.com/SEU-VIPGroup/FG-BMK. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_14988 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation Yu, Hong-Tao Peng, Yuxin Belongie, Serge Wei, Xiu-Shen Computer Vision and Pattern Recognition Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal perception capabilities, garnering significant attention. While numerous evaluation studies have emerged, assessing LVLMs both holistically and on specialized tasks, fine-grained image tasks-fundamental to computer vision-remain largely unexplored. To fill this gap, we introduce a comprehensive fine-grained evaluation benchmark, i.e., FG-BMK, comprising 1.01 million questions and 0.33 million images. Our evaluation systematically examines LVLMs from both human-oriented and machine-oriented perspectives, focusing on their semantic recognition and fine-grained feature representation capabilities. Through extensive experiments on twelve representative LVLMs/VLMs, we uncover key findings regarding the influence of training paradigms, modality alignment, perturbation susceptibility, and fine-grained category reasoning on task performance. This work provides critical insights into the limitations of current LVLMs and offers guidance for future data construction and model design in the development of more advanced LVLMs. Our code is open-source and available at https://github.com/SEU-VIPGroup/FG-BMK. |
| title | Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.14988 |