Evaluating Model Perception of Color Illusions in Photorealistic Scenes
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913602423750656 |
|---|---|
| author | Mao, Lingjun Tang, Zineng Suhr, Alane |
| author_facet | Mao, Lingjun Tang, Zineng Suhr, Alane |
| contents | We study the perception of color illusions by vision-language models. Color illusion, where a person's visual system perceives color differently from actual color, is well-studied in human vision. However, it remains underexplored whether vision-language models (VLMs), trained on large-scale human data, exhibit similar perceptual biases when confronted with such color illusions. We propose an automated framework for generating color illusion images, resulting in RCID (Realistic Color Illusion Dataset), a dataset of 19,000 realistic illusion images. Our experiments show that all studied VLMs exhibit perceptual biases similar human vision. Finally, we train a model to distinguish both human perception and actual pixel differences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_06184 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluating Model Perception of Color Illusions in Photorealistic Scenes Mao, Lingjun Tang, Zineng Suhr, Alane Computer Vision and Pattern Recognition We study the perception of color illusions by vision-language models. Color illusion, where a person's visual system perceives color differently from actual color, is well-studied in human vision. However, it remains underexplored whether vision-language models (VLMs), trained on large-scale human data, exhibit similar perceptual biases when confronted with such color illusions. We propose an automated framework for generating color illusion images, resulting in RCID (Realistic Color Illusion Dataset), a dataset of 19,000 realistic illusion images. Our experiments show that all studied VLMs exhibit perceptual biases similar human vision. Finally, we train a model to distinguish both human perception and actual pixel differences. |
| title | Evaluating Model Perception of Color Illusions in Photorealistic Scenes |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.06184 |