analiza aiqbench
This study benchmarks Vision Transformer foundation models for unsupervised clustering of animal images, demonstrating that DINOv3 embeddings can achieve near-manual annotation accuracy while drastically reducing the need for expert labeling in ecological research. By evaluating multiple dimensionality reduction and clustering algorithms across 60 species, the authors provide an open-source toolkit and practical guidelines for applying these AI methods to biodiversity monitoring tasks.
Vision Transformers for Zero-Shot Clustering of Animal Images: A Comparative Benchmarking Study
De la arXiv
Text integralarXiv:2602.03894v2 Announce Type: replace-cross Abstract: Manual labelling of animal images remains a significant bottleneck in ecological research, limiting the scale and efficiency of biodiversity monitoring efforts. This study investigates whether state-of-the-art Vision Transformer (ViT) foundation models can reduce thousands of unlabelled animal images directly to species-level clusters. We present a comprehensive benchmarking framework evaluating five ViT models combined with five dimensionality reduction techniques and four clustering algorithms, two supervised and two unsupervised, across 60 species (30 mammals and 30 birds), with each test using a random subset of 200 validated images per species.
We investigate when clustering succeeds at species-level, where it fails, and whether clustering within the species-level reveals ecologically meaningful patterns such as sex, age, or phenotypic variation. Our primary, fully unsupervised finding is that DINOv3 embeddings with t-SNE and HDBSCAN reach a species-level V-measure of 0.943 with no prior knowledge of the number of species, rejecting only 1.14% of images as outliers requiring expert review. Assigning one label per cluster reduces manual labelling actions by 98.3%, with 94.6% of clustered images labelled correctly, close to the accuracy of manual annotation.
When the true number of species is supplied, supervised hierarchical clustering raises this to 0.958, which we report as an upper bound as the species count is typically unknown in real deployments. We further demonstrate robustness to realistic long-tailed distributions of species and show that intentional over-clustering can reveal intra-specific variation including age classes, sexual dimorphism, and pelage differences. We introduce an open-source benchmarking toolkit and provide recommendations for ecologists to select appropriate methods for sorting their specific taxonomic groups and data.
Abstract:Manual labelling of animal images remains a significant bottleneck in ecological research, limiting the scale and efficiency of biodiversity monitoring efforts. This study investigates whether state-of-the-art Vision Transformer (ViT) foundation models can reduce thousands of unlabelled animal images directly to species-level clusters. We present a comprehensive benchmarking framework evaluating five ViT models combined with five dimensionality reduction techniques and four clustering algorithms, two supervised and two unsupervised, across 60 species (30 mammals and 30 birds), with each test using a random subset of 200 validated images per species.
We investigate when clustering succeeds at species-level, where it fails, and whether clustering within the species-level reveals ecologically meaningful patterns such as sex, age, or phenotypic variation. Our primary, fully unsupervised finding is that DINOv3 embeddings with t-SNE and HDBSCAN reach a species-level V-measure of 0.943 with no prior knowledge of the number of species, rejecting only 1.14% of images as outliers requiring expert review. Assigning one label per cluster reduces manual labelling actions by 98.3%, with 94.6% of clustered images labelled correctly, close to the accuracy of manual annotation.
When the true number of species is supplied, supervised hierarchical clustering raises this to 0.958, which we report as an upper bound as the species count is typically unknown in real deployments. We further demonstrate robustness to realistic long-tailed distributions of species and show that intentional over-clustering can reveal intra-specific variation including age classes, sexual dimorphism, and pelage differences. We introduce an open-source benchmarking toolkit and provide recommendations for ecologists to select appropriate methods for sorting their specific taxonomic groups and data.
Published version (open access, CC BY 4.0): IET Computer Vision 20(1), e70091 (2026)
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2602.03894 [cs.CV]
(or arXiv:2602.03894v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2602.03894 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.1049/cvi2.70091 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 381 cuvinte
Subiecte conexe
BanglaRhet: Benchmarking Classical and Transformer Models for Rhetorical and Persuasion Detection in Bangla Political Speech
arXiv · 8 oct. 2026
David vs. Goliath in Next Activity Prediction: Argmax vs. LSTM, Transformer, and LLM
arXiv · 16 iun. 2026
StabilityBench: Benchmarking Instability in LLMs
arXiv · 24 iul. 2026
MetaPI: Constructing Prompt Injection Benchmarks from LLM Agent Benchmark
arXiv · 10 nov. 2026