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

Vaibhav Rathore; Divyam Gupta; Moloud Abdar; Subhasis Chaudhuri; Biplab Banerjee

cs.CVMar 15, 2026

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

Authors: Vaibhav Rathore, Divyam Gupta, Moloud Abdar, Subhasis Chaudhuri, Biplab Banerjee

Organizations: Indian Institute of Technology Bombay, Mumbai, India. · Northeastern University, Boston, USA. · The University of Queensland, Brisbane, Australia.

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.]

Figures & tables

Figure 1: FG-DG-GCD extends DG-GCD Rathore et al. (2025a) to the fine-grained regime, where a model trained on labeled source classes must simultaneously recognize known and discover novel categories in unseen, distribution-shifted domains. Compared with coarse-grained DG-GCD Rathore et al. (2025a) ; Rathore et al. (2025b) , this setting is substantially harder due to subtle inter-class cues, high intra-class similarity, and stronger sensitivity to domain-induced appearance changes. FoCUS addresses this challenge through part-consistent representation learning and uncertainty-aware open-space regularization, outperforming existing GCD, FG-GCD, and DG-GCD baselines on the proposed benchmarks (Table 4 ).

Method	Fine-	Domain	Novel Class	Design Choice		Part-
	Grained	Shift	Discovery	Synthetic-Free	Episodic-Free	Aware
DG Zhou et al. (2021a) ; Wang et al. (2023) ; Xu et al. (2023)	✗	✓	✗	✓	✓	✗
FG-DG Yu et al. (2025) ; Wang et al. (2026)	✓	✓	✗	✓	✓	✓
(FG)GCD Vaze et al. (2022) ; Pu et al. (2023) ; Zhang et al. (2023)	✓	✗	✓	✓	✓	✗
DG-GCD Rathore et al. (2025a) ; Rathore et al. (2025b)	✗	✓	✓	✗	✗	✗
FoCUS (Ours)	✓	✓	✓	✓	✓	✓

Table 1: FoCUS is the first framework to jointly address fine-grained recognition , domain generalization , and novel class discovery within a single formulation. Unlike prior DG-GCD methods Rathore et al. (2025a) ; Rathore et al. (2025b) , it is synthetic-free and episodic-free , reducing training FLOPs by 3

\times

and 300

\times

compared with HiDISC Rathore et al. (2025b) and DG 2 CD-Net Rathore et al. (2025a) , respectively, while delivering stronger generalization and discovery performance.

Dataset	#Classes	Split	#Images			#Domains	FID Scores ( $\downarrow$ )
Dataset	#Classes	K:N	Real	Painting	Sketch	#Domains	R $\rightarrow$ P	R $\rightarrow$ S	S $\rightarrow$ P
CUB-200-2011-MD	200	100:100	11,788	11,788	11,788	3	55.13	160.18	106.95
Stanford Cars-MD	196	98:98	16,365	16,365	16,365	3	29.62	62.86	41.63
FGVC-Aircraft-MD	100	50:50	10,000	10,000	10,000	3	64.33	170.27	81.78

Table 2: Summary of the tri-domain fine-grained benchmarks used in this study. Each dataset comprises real photographs and their painting and sketch counterparts synthesized via SDXL with T2I-Adapter. FID Eiter et al. (1994) scores measure cross-domain shifts, reflecting the realism and diversity of the generated domains.

Dataset	Domain Pair	LPIPS (AlexNet)	LPIPS (VGG)
CUB-200-2011-MD	Real $\rightarrow$ Sketch	0.5019	0.5558
CUB-200-2011-MD	Real $\rightarrow$ Painting	0.5564	0.5801
FGVC-Aircraft-MD	Real $\rightarrow$ Sketch	0.4525	0.5441
FGVC-Aircraft-MD	Real $\rightarrow$ Painting	0.5158	0.5585
Stanford Cars-MD	Real $\rightarrow$ Sketch	0.4284	0.4718
Stanford Cars-MD	Real $\rightarrow$ Painting	0.4599	0.4784

Table 3: LPIPS distances between real and synthetic domains. Lower values indicate higher perceptual similarity. Scores are computed with AlexNet and VGG backbones.

Figure 2: Overview of FoCUS . A ViT encoder produces global [CLS] and patch-level features. The Domain-Consistent Parts Discovery (DCPD) module extracts geometry-stable part features $\mathbf{f}_{\text{part}}$ , which are fused with the global representation $\mathbf{f}_{\text{cls}}$ and optimized using InfoNCE ( $\mathcal{L}_{\text{InfoNCE}}$ ) and supervised contrastive ( $\mathcal{L}_{\text{SCon}}$ ) losses for fine-grained discrimination. In parallel, the Uncertainty-Aware Feature Augmentation (UFA) module uses EMA-based class statistics $(\bm{\mu}_{y},\bm{\Sigma}_{y})$ to synthesize outlier features $\widetilde{\mathbf{\mathcal{Z}}}$ from class-tail, between-class, and hypersphere regions. A shared classifier $g(\cdot)$ is trained on real features using cross-entropy, and on synthetic outliers using energy-based outlier exposure ( $\mathcal{L}_{\text{OE}}$ ) and entropy maximization ( $\mathcal{L}_{\text{ENT}}$ ). Together, these two pathways yield discriminative, robust, and calibrated embeddings for category discovery under unseen domain shift.

Figure 3: Domain-Consistent Attention across Domains. Compared to prior part-localization methods (FOCUS+APL Dai et al. (2025) ), our DCPD module maintains stable attention on the same geometric regions (e.g., bird’s head) across domain shifts from Real $\rightarrow$ Painting $\rightarrow$ Sketch , demonstrating strong domain-invariant part discovery. See Sup. Mat. for KL divergence between the cross-domain attention maps for quantitative validation.

Figure 4: Progressive Attention Refinement within DCPD . Left to Right,the evolution of : (1) Self-Attention Priors provide initial coarse localization; (2) Image-Conditioned Part Queries refine attention to align with instance-specific geometries; (3) Differentiable Patch-to-Part Assignment creates sharp, spatially exclusive part delineations; and finally, (4) Part-Global Fusion integrates these precise structural cues into a unified representation.

Figure 5: t-SNE Van der Maaten and Hinton (2008) of DCPD embeddings with source (colored) and UFA outliers (black) occupying low-density boundaries, modeling open-space uncertainty and improving calibration.

Methods	Venue	CUB-200-2011-MD			Stanford Cars-MD			FGVC-Aircraft-MD			Average
Methods	Venue	All	Old	New	All	Old	New	All	Old	New	All	Old	New
GCD Baselines
ViT Dosovitskiy et al. (2021)	ICLR’21	24.17	24.88	23.53	16.04	15.40	17.11	17.37	16.62	18.18	19.19	18.97	19.60
GCD Vaze et al. (2022)	CVPR’22	33.34	36.11	30.51	36.46	34.61	38.89	18.95	18.49	20.01	29.58	29.74	29.80
SimGCD Wen et al. (2023)	ICCV’23	29.61	32.40	26.63	40.90	34.61	47.12	33.69	30.50	36.88	34.73	32.50	36.88
CMS Choi et al. (2024)	CVPR’24	7.82	7.63	8.00	4.79	4.71	4.90	7.11	7.17	7.05	6.57	6.50	6.65

Table 4: Performance comparison across three fine-grained datasets averaged over all domain combinations. Baselines are grouped by category (GCD, FG-GCD, FG-DG, DG-GCD and Part Aware Baselines). Our method consistently outperforms prior works, both with and without synthetic domain augmentation.

\Delta

quantifies our method’s gain or drop relative to the best prior baseline. ( Bold : best, underline : second-best).

Figure 6: t-SNE Van der Maaten and Hinton (2008) visualizations of the target domain (“ Photo ”) clusters, as produced by GCD Vaze et al. (2022) , SelEx Rastegar et al. (2024) , DG 2 CD-Net Rathore et al. (2025a) , HiDISC Rathore et al. (2025b) , and FoCUS (Ours) on the CUB-200-2011-MD dataset, with “ Sketch ” as the source domain. FoCUS produces a clean and compact embedding space, clearly separating the target clusters. For visualization, we randomly select 15 classes and 20 samples per class.

Method	Venue	PACS			Office-Home			DomainNet			Avg.
		All	Old	New	All	Old	New	All	Old	New	All	Old	New
ViT Dosovitskiy et al. (2021)	ICLR’21	41.98	50.91	33.16	26.17	29.13	21.62	25.35	26.48	22.41	31.17	35.51	25.73
GCD Vaze et al. (2022)	CVPR’22	52.28	62.20	38.39	52.71	54.19	50.29	27.41	27.88	26.13	44.13	48.09	38.27
SimGCD Wen et al. (2023)	ICCV’23	34.55	38.64	30.51	36.32	49.48	13.55	2.84	2.16	3.75	24.57	30.09	15.94
CMS Choi et al. (2024)	CVPR’24	28.95	28.13	36.80	10.02	9.66	10.53	2.33	2.40	2.17	13.77	13.40	16.50
SelfEx Rastegar et al. (2024)	ECCV’24	71.82	73.37	71.55	50.18	48.59	52.16	24.78	24.99	24.21	48.93	48.98	49.31

Table 5: State-of-the-Art Clustering Accuracy (%) on Domain Generalization Benchmarks. This table details the clustering accuracy for known (Old), novel (New), and overall (All) categories across the PACS, Office-Home, and DomainNet datasets. Our proposed FoCUS consistently achieves leading performance, surpassing both synthetic domain augmentation and non-synthetic baselines. This is particularly significant as FoCUS uses substantially fewer synthetic domains (down to 2 from 6/9) or even no synthetic data at all, highlighting its efficiency and effectiveness. ( Bold : best, underline : second best).

Config.	Loss Components					Accuracy (%) $\uparrow$
Config.	$\mathcal{L}_{\mathrm{SCon}}$	$\mathcal{L}_{\mathrm{InfoNCE}}$	$\mathcal{L}_{\mathrm{CE}}$	$\mathcal{L}_{\mathrm{OE}}$	$\mathcal{L}_{\mathrm{ENT}}$	All	Old	New
C-0	✗	✗	✗	✗	✗	16.04	15.40	17.11
C-1	✓	✗	✗	✗	✗	28.58	26.92	30.78
C-2	✗	✓	✗	✗	✗	9.46	9.01	10.06
C-3	✓	✓	✗	✗	✗	31.64	30.32	33.39
C-4	✓	✓	✓	✗	✗	36.22	33.36	39.99

Table 6: Ablation on loss components. Evaluation on Stanford Cars-MD.

Table 7: Ablation studies on the Stanford Cars-MD dataset: (a) feature fusion, (b) optimal parts

T

in DCPD, (c) prior selection percentile.

Table 8: Ablation results on the Stanford Cars-MD dataset. (a) Impact of UFA sampler. (b) Sensitivity to Margin

m

. (c) Synthetic OOD features.

Method	Space	Mechanism	Open-Space	All	Old	New
MixUp Zhang et al. (2017)	Feature	Linear interp.	✗	36.85	34.20	39.12
CutMix Yun et al. (2019)	Image	Patch repl.	✗	37.42	34.91	40.58
Manifold Verma et al. (2019)	Feature	Manifold blend.	✗	38.05	35.44	41.32
VAT Miyato et al. (2018)	Feature	Adv. perturb.	Partial	39.88	36.70	43.21
OE Hendrycks et al. (2018)	Ext. data	OOD superv.	✓	41.02	37.91	45.15
UFA (Ours)	Feature	Uncert.-driven	Self-gen.	43.36	38.55	48.12

Table 9: UFA comparison. Uncertainty-guided feature-space sampling versus baselines.

Figure 8: Sample images from the Stanford Cars-MD benchmark, illustrating diversity in make, model, and viewpoint. The benchmark contains $(16{,}185\times 3)$ images over 196 car categories across the real , painting , and sketch domains.

Figure 9: Example images from the FGVC-Aircraft-MD benchmark, showing variation across aircraft types, manufacturers, and liveries. The benchmark includes $(10{,}000\times 3)$ images spanning 100 aircraft variants, each characterized by subtle structural and texture cues.

Figure 10: Representative samples from the CUB-200-2011-MD benchmark, containing $(11{,}788\times 3)$ images from 200 bird species across three domains. Its high category count and fine visual granularity make it a particularly challenging benchmark for FG-DG-GCD.

Feature	APL Dai et al. (2025)	DCPD (Ours)
Target Task	Standard GCD	Fine-Grained Domain-Generalized GCD
Query Paradigm	Global & Static ( $\mathbf{Q}$ )	Adaptive & Instance-Conditioned ( $\mathbf{Q}_{I}$ )
Discovery Prior	Perceptual (DINO features)	Geometric Blueprint (Thresholded Masks)
Global Context	Discarded (Part-Replacement)	Preserved (Dual-Stream Fusion)
Domain Handling	Domain-Agnostic	Explicit Domain-Aware Architectural Bias

Table 11: Architectural and functional comparison between APL and the proposed DCPD.

Variant	All ( $\uparrow$ )	PUE ( $\downarrow$ )
APL Dai et al. (2025)	34.70	1.95
DCPD	43.36	1.72
+ Div. Loss	43.19	1.83

Table 12: DCPD vs. APL.

Method	R→P	R→S	P→S
FOCUS+APL Dai et al. (2025)	0.274	0.312	0.295
DCPD (Ours)	0.143	0.167	0.159

Table 13: Cross-Domain Attention Consistency (KL

\downarrow

). Lower values indicate more stable and domain-invariant attention maps. We report KL divergence between domains for Real

\rightarrow

Painting, Real

\rightarrow

Sketch, and Painting

\rightarrow

Sketch.

Method	CUB	SCARS	Aircraft
Ground Truth	200	196	100
DG 2 CD-Net	285	268	175
HiDISC	245	230	150
FoCUS	230	180	125

Table 14: Estimated number of clusters . All methods deviate from the ground-truth number of categories. Our method consistently achieves the smallest error (green), while other methods show moderate (orange) or large (red) deviations.

CUB-200-2011-MD
Methods	Painting $\rightarrow$ Sketch			Painting $\rightarrow$ Real			Real $\rightarrow$ Sketch			Real $\rightarrow$ Painting			Sketch $\rightarrow$ Real			Sketch $\rightarrow$ Painting
Methods	All	Old	New	All	Old	New	All	Old	New	All	Old	New	All	Old	New	All	Old	New
ViT Dosovitskiy et al. (2021)	37.44	36.27	38.62	10.48	11.23	9.82	28.59	29.73	27.32	12.27	12.74	11.86	25.29	25.32	25.27	30.96	34.01	28.28
GCD Vaze et al. (2022)	47.43	51.14	43.71	17.64	18.52	16.86	35.84	40.44	31.92	16.32	17.36	15.40	33.54	34.28	32.84	43.83	45.56	42.30
CMS Choi et al. (2024)	6.47	6.03	6.85	5.32	5.34	5.29	11.30	11.24	11.35	5.60	5.42	5.77	12.13	11.52	12.70	6.11	6.21	6.02
SimGCD Wen et al. (2023)	37.34	40.69	34.02	13.03	15.23	10.84	39.22	44.82	32.67	19.39	20.65	18.15	29.05	30.82	26.98	39.63	42.20	37.09

Table 25

Stanford-Cars-MD
Methods	Painting $\rightarrow$ Sketch			Painting $\rightarrow$ Real			Real $\rightarrow$ Sketch			Real $\rightarrow$ Painting			Sketch $\rightarrow$ Real			Sketch $\rightarrow$ Painting
Methods	All	Old	New	All	Old	New	All	Old	New	All	Old	New	All	Old	New	All	Old	New
ViT Dosovitskiy et al. (2021)	1.17	0.00	2.72	22.70	21.76	23.93	27.39	27.33	27.47	19.10	18.24	20.23	24.68	23.99	25.58	1.17	1.10	2.72
GCD Vaze et al. (2022)	44.24	41.80	47.45	32.69	30.94	35.00	36.80	35.19	38.92	26.49	24.71	28.83	36.47	35.07	38.32	41.61	40.05	43.66
CMS Choi et al. (2024)	5.11	5.13	5.09	4.53	4.46	4.62	4.85	4.87	4.84	4.79	4.34	5.37	4.85	4.89	4.81	4.62	4.58	4.68
SimGCD Wen et al. (2023)	48.16	40.53	55.73	39.18	32.49	45.80	42.23	36.32	48.08	32.51	27.34	37.62	38.90	32.97	44.77	44.41	38.03	50.73

Table 26

FGVC-Aircraft-MD
Methods	Painting $\rightarrow$ Sketch			Painting $\rightarrow$ Real			Real $\rightarrow$ Sketch			Real $\rightarrow$ Painting			Sketch $\rightarrow$ Real			Sketch $\rightarrow$ Painting
Methods	All	Old	New	All	Old	New	All	Old	New	All	Old	New	All	Old	New	All	Old	New
ViT Dosovitskiy et al. (2021)	17.45	18.22	16.60	16.10	14.07	18.29	21.13	18.97	23.47	18.41	18.40	18.41	14.75	14.88	14.61	16.37	15.17	17.67
GCD Vaze et al. (2022)	37.14	34.20	40.32	2.01	3.86	1.10	31.47	29.18	33.96	2.01	3.86	1.15	2.01	3.86	1.20	39.03	35.99	42.32
CMS Choi et al. (2024)	7.49	7.50	7.49	7.10	5.94	8.36	6.80	6.75	6.87	6.83	7.09	6.55	6.98	7.96	5.93	7.46	7.79	7.12
SimGCD Wen et al. (2023)	23.59	20.04	27.15	33.60	32.04	35.15	38.32	35.95	40.71	29.92	24.55	35.25	39.03	35.11	42.95	37.68	35.28	40.07

Table 27

Panel of Figure 12 — Figure 12: Comparison of InfoNCE Loss convergence across the three source domains on the CUB-200-2011-MD dataset.

Panel of Figure 13 — Figure 13: Comparison of Sup_Con Loss convergence across the three source domains on the CUB-200-2011-MD dataset.

Panel of Figure 14 — Figure 14: Comparison of CLS Loss convergence across the three source domains on the CUB-200-2011-MD dataset.

Panel of Figure 15 — Figure 15: Comparison of ENT Loss convergence across the three source domains on the CUB-200-2011-MD dataset.

Panel of Figure 16 — Figure 16: Comparison of OE Loss convergence across the three source domains on the CUB-200-2011-MD dataset.

Generalized Category Discovery

Generalized category discovery learns known classes from labeled examples while grouping unlabeled data into known and previously unseen categories.

See all 25 papers

Apr 29, 2026

Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models

Generalized Category Discovery (GCD) aims to categorize unlabelled instances from both known and unknown classes by transferring knowledge from labelled data of known classes. Existing methods assume all data comes from a single domain, yet real-world unlabelled data often exhibits domain shifts alongside semantic shifts. We study GCD under domain shifts and propose three frameworks that adapt foundation models, ranging from self-supervised vision models to vision-language models. (i) HiLo disentangles domain and semantic features through multi-level feature extraction and mutual information minimization, combined with PatchMix augmentation and curriculum sampling. (ii) HLPrompt extends HiLo with semantic-aware spatial prompt tuning to suppress background and domain noise. (iii) VLPrompt leverages vision-language models via factorized textual prompts and cross-modal consistency regularization. The three methods share core design principles while operating on different foundation backbones, making them suitable for different deployment scenarios. Extensive experiments on synthetic corruptions and real-world multi-domain shifts demonstrate consistent improvements over strong baselines. Project page: https://visual-ai.github.io/hilo/

Hongjun Wang, Po Hu, Kai Han

School of Computing and Data Science, The University of Hong Kong

VLM Adaptation Vision Foundation Model Adaptation Generalized Category Discovery

Ask PDF

Aug 26, 2026

CloSeR: Unified Relational Distillation from Closed-Set Teachers for Category Discovery

Generalized Category Discovery (GCD) is an intriguing open-world problem that has garnered increasing attention: given partially labelled data, the goal is to correctly recognize known classes while discovering coherent novel categories from unlabelled samples. Recent GCD methods typically adapt foundation models by jointly optimizing supervised classification and unsupervised discovery objectives on mixed labelled and unlabelled data. While effective, this coupled training can entangle closed-set recognition and open-set discovery, leading to objective conflict and biased predictions, and may disturb the semantic geometry of pretrained representations under limited labels and noisy pseudo-labels. We propose CloSeR, a simple plug-and-play framework that injects Closed-Set Relational knowledge into GCD training. CloSeR first builds a domain-adapted closed-set teacher by tuning lightweight block-wise adapters on labelled known-class data while keeping the foundation model backbone frozen, thereby preserving pretrained priors at low training cost. It then transfers the teacher's knowledge to downstream GCD via Unified Relational Distillation (URD), which distills complementary global sample-to-prototype relations to anchor known-class semantics and local sample-to-sample relations to preserve neighborhood structure, using separate feature pathways to reduce optimization interference. CloSeR is head-agnostic and readily integrates with both parametric and non-parametric GCD methods. Extensive experiments with DINO and DINOv2 backbones on six benchmarks (CIFAR-10/100, ImageNet-100, CUB, Stanford-Cars, and FGVC-Aircraft) show consistent gains over GCD baselines, achieving state-of-the-art performance. Project page: https://visual-ai.github.io/closer/

Yuanpei Liu, Zhenqi He, Jialu Tang +1

Visual AI Lab, The University of Hong Kong, Hong Kong SAR

Generalized Category Discovery Relational Knowledge Distillation

Ask PDF

Sep 9, 2026

CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization

Generalized Category Discovery (GCD) assigns unlabeled instances, mixed with labeled data, to known or novel categories, requiring human-like compositional reasoning: reusing primitives learned from known classes and deciding when new combinations imply new categories. Existing GCD methods operate on unstructured token features and struggle to extrapolate to novel compositions. We propose CoGe-GCD, which rethinks GCD through compositional generalization with two coupled stages. (i) Compositional Perception structures patch tokens by mapping them to a small vocabulary of primitives and refining token embeddings via competitive token-primitive assignment and information passing, yielding coherent groups for discovery. (ii) Generalizing Induction exploits the induced geometric structure and applies a structure-preserving calibration over spatial relations, maintaining probabilistic semantics while improving extrapolation to unseen primitive combinations. CoGe-GCD is implemented as an inductive-bias module between backbone and projection head, without modifying heads or losses, and can be plugged into diverse GCD frameworks. On standard benchmarks, it consistently improves all-class accuracy, unknown-class number estimation, and geometric quality, with marginal computational overhead. Code is available at https://github.com/lytang63/CoGe-GCD.

Luyao Tang, Jiewei Zheng, Kunze Huang +3

The University of Hong Kong, Hong Kong SAR, China · Xiamen University, Xiamen, China · Shenzhen University, Shenzhen, China.

Geometric Representation Learning Classification Compositional Generalization Generalized Category Discovery

Ask PDF

Abstract

Generalized Category Discovery

Explore similar work

Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models

CloSeR: Unified Relational Distillation from Closed-Set Teachers for Category Discovery

CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization