TaxaDiffusion: Progressively Trained Diffusion Model for Fine-Grained Species Generation

2025-06-02

AI Summary

Fine-grained animal image generation suffers from low morphological and identity fidelity, poor generalization under few-shot settings, and neglect of biologically grounded phylogenetic relationships among species. Method: We propose the first taxonomy-aware progressive diffusion framework guided by the biological classification hierarchy (family–genus–species), incorporating multi-level taxonomy-conditioned encoding and a stage-wise hierarchical training strategy to enable cross-level knowledge transfer and fine-grained semantic alignment. Contribution/Results: Evaluated on three fine-grained animal benchmarks with ≤10 images per class, our method significantly outperforms existing approaches, achieving state-of-the-art performance in image fidelity, species recognition accuracy, and cross-species consistency. This demonstrates the effectiveness and scalability of integrating biological priors—specifically taxonomic structure—into generative modeling.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Natural Language Processing: Generation

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We propose TaxaDiffusion, a taxonomy-informed training framework for diffusion models to generate fine-grained animal images with high morphological and identity accuracy. Unlike standard approaches that treat each species as an independent category, TaxaDiffusion incorporates domain knowledge that many species exhibit strong visual similarities, with distinctions often residing in subtle variations of shape, pattern, and color. To exploit these relationships, TaxaDiffusion progressively trains conditioned diffusion models across different taxonomic levels -- starting from broad classifications such as Class and Order, refining through Family and Genus, and ultimately distinguishing at the Species level. This hierarchical learning strategy first captures coarse-grained morphological traits shared by species with common ancestors, facilitating knowledge transfer before refining fine-grained differences for species-level distinction. As a result, TaxaDiffusion enables accurate generation even with limited training samples per species. Extensive experiments on three fine-grained animal datasets demonstrate that outperforms existing approaches, achieving superior fidelity in fine-grained animal image generation. Project page: https://amink8.github.io/TaxaDiffusion/

Problem

• Generates fine-grained animal images with high accuracy
• Incorporates taxonomy to exploit visual similarities among species
• Uses hierarchical training to improve species-level distinction

Innovation

• Taxonomy-informed diffusion model training
• Progressive hierarchical species-level refinement
• Leverages visual similarities for accurate generation