FOCUS: Bridging Fine-Grained Recognition and Open-World Discovery across Domains

2026-03-15

AI Summary

This work addresses the dual challenge of recognizing known classes and discovering novel unknown classes in fine-grained cross-domain scenarios by introducing, for the first time, the Fine-Grained Domain-Generalized Generalized Category Discovery (FG-DG-GCD) task and establishing a corresponding benchmark. To tackle this problem, the authors propose FoCUS, a unified single-stage framework that integrates Domain-Consistent Part Discovery (DCPD) and Uncertainty-Aware Feature Augmentation (UFA), complemented by diffusion adapter-based style transfer, fine-grained clustering, and feature regularization to effectively mitigate domain shift and capture subtle inter-class distinctions. Experimental results demonstrate that FoCUS significantly outperforms existing GCD, FG-GCD, and DG-GCD baselines on the proposed benchmark by 3.28%, 9.68%, and 2.07% in clustering accuracy, respectively, while achieving nearly three times higher computational efficiency and maintaining competitive performance on coarse-grained tasks.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Diffusion Models for Vision
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

We introduce the first unified framework for *Fine-Grained Domain-Generalized Generalized Category Discovery* (FG-DG-GCD), bringing open-world recognition closer to real-world deployment under domain shift. Unlike conventional GCD, which assumes labeled and unlabeled data come from the same distribution, DG-GCD learns only from labeled source data and must both recognize known classes and discover novel ones in unseen, unlabeled target domains. This problem is especially challenging in fine-grained settings, where subtle inter-class differences and large intra-class variation make domain generalization significantly harder. To support systematic evaluation, we establish the first *FG-DG-GCD benchmarks* by creating identity-preserving *painting* and *sketch* domains for CUB-200-2011, Stanford Cars, and FGVC-Aircraft using controlled diffusion-adapter stylization. On top of this ,we propose FoCUS, a single-stage framework that combines *Domain-Consistent Parts Discovery* (DCPD) for geometry-stable part reasoning with *Uncertainty-Aware Feature Augmentation* (UFA) for confidence-calibrated feature regularization through uncertainty-guided perturbations. Extensive experiments show that FoCUS outperforms strong GCD, FG-GCD, and DG-GCD baselines by **3.28%**, **9.68%**, and **2.07%**, respectively, in clustering accuracy on the proposed benchmarks. It also remains competitive on coarse-grained DG-GCD tasks while achieving nearly **3x** higher computational efficiency than the current state of the art. ^[Code and datasets will be released upon acceptance.]

Problem

• Fine-Grained Recognition
• Domain Generalization
• Generalized Category Discovery
• Open-World Recognition
• Domain Shift

Innovation

• Fine-Grained Domain-Generalized Generalized Category Discovery
• Domain-Consistent Parts Discovery
• Uncertainty-Aware Feature Augmentation
• Open-World Recognition
• Cross-Domain Benchmark