SFOOD: A Multimodal Benchmark for Comprehensive Food Attribute Analysis Beyond RGB with Spectral Insights
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915373408845824 |
|---|---|
| author | Xu, Zhenbo Yang, Jinghan Huang, Gong Feng, Jiqing Liu, Liu Sun, Ruihan Meng, Ajin Zhang, Zhuo He, Zhaofeng |
| author_facet | Xu, Zhenbo Yang, Jinghan Huang, Gong Feng, Jiqing Liu, Liu Sun, Ruihan Meng, Ajin Zhang, Zhuo He, Zhaofeng |
| contents | With the rise and development of computer vision and LLMs, intelligence is everywhere, especially for people and cars. However, for tremendous food attributes (such as origin, quantity, weight, quality, sweetness, etc.), existing research still mainly focuses on the study of categories. The reason is the lack of a large and comprehensive benchmark for food. Besides, many food attributes (such as sweetness, weight, and fine-grained categories) are challenging to accurately percept solely through RGB cameras. To fulfill this gap and promote the development of intelligent food analysis, in this paper, we built the first large-scale spectral food (SFOOD) benchmark suite. We spent a lot of manpower and equipment costs to organize existing food datasets and collect hyperspectral images of hundreds of foods, and we used instruments to experimentally determine food attributes such as sweetness and weight. The resulting benchmark consists of 3,266 food categories and 2,351 k data points for 17 main food categories. Extensive evaluations find that: (i) Large-scale models are still poor at digitizing food. Compared to people and cars, food has gradually become one of the most difficult objects to study; (ii) Spectrum data are crucial for analyzing food properties (such as sweetness). Our benchmark will be open source and continuously iterated for different food analysis tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_04412 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SFOOD: A Multimodal Benchmark for Comprehensive Food Attribute Analysis Beyond RGB with Spectral Insights Xu, Zhenbo Yang, Jinghan Huang, Gong Feng, Jiqing Liu, Liu Sun, Ruihan Meng, Ajin Zhang, Zhuo He, Zhaofeng Computer Vision and Pattern Recognition With the rise and development of computer vision and LLMs, intelligence is everywhere, especially for people and cars. However, for tremendous food attributes (such as origin, quantity, weight, quality, sweetness, etc.), existing research still mainly focuses on the study of categories. The reason is the lack of a large and comprehensive benchmark for food. Besides, many food attributes (such as sweetness, weight, and fine-grained categories) are challenging to accurately percept solely through RGB cameras. To fulfill this gap and promote the development of intelligent food analysis, in this paper, we built the first large-scale spectral food (SFOOD) benchmark suite. We spent a lot of manpower and equipment costs to organize existing food datasets and collect hyperspectral images of hundreds of foods, and we used instruments to experimentally determine food attributes such as sweetness and weight. The resulting benchmark consists of 3,266 food categories and 2,351 k data points for 17 main food categories. Extensive evaluations find that: (i) Large-scale models are still poor at digitizing food. Compared to people and cars, food has gradually become one of the most difficult objects to study; (ii) Spectrum data are crucial for analyzing food properties (such as sweetness). Our benchmark will be open source and continuously iterated for different food analysis tasks. |
| title | SFOOD: A Multimodal Benchmark for Comprehensive Food Attribute Analysis Beyond RGB with Spectral Insights |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.04412 |