We introduce Fish-Visual Trait Analysis (Fish-Vista), the first organismal image dataset designed for the analysis of visual traits of aquatic species directly from images using problem formulations in computer vision. Fish-Vista contains 69,126 annotated images spanning 4,154 fish species, curated and organized to serve three downstream tasks of species classification, trait identification, and trait segmentation. Our work makes two key contributions. First, we perform a fully reproducible data processing pipeline to process images sourced from various museum collections. We annotate these images with carefully curated labels from biological databases and manual annotations to create an AI-ready dataset of visual traits, contributing to the advancement of AI in biodiversity science. Second, our proposed downstream tasks offer fertile grounds for novel computer vision research in addressing a variety of challenges such as long-tailed distributions, out-of-distribution generalization, learning with weak labels, explainable AI, and segmenting small objects. We benchmark the performance of several existing methods for our proposed tasks to expose future research opportunities in AI for biodiversity science problems involving visual traits.
View on arXiv@article{mehrab2025_2407.08027,
title={ Fish-Vista: A Multi-Purpose Dataset for Understanding & Identification of Traits from Images },
author={ Kazi Sajeed Mehrab and M. Maruf and Arka Daw and Abhilash Neog and Harish Babu Manogaran and Mridul Khurana and Zhenyang Feng and Bahadir Altintas and Yasin Bakis and Elizabeth G Campolongo and Matthew J Thompson and Xiaojun Wang and Hilmar Lapp and Tanya Berger-Wolf and Paula Mabee and Henry Bart and Wei-Lun Chao and Wasila M Dahdul and Anuj Karpatne },
journal={arXiv preprint arXiv:2407.08027},
year={ 2025 }
}