Dual Feature Decoupling for Fine-Grained OOD Detection
Quick Answer
This paper shows that The Dual Feature Decoupling Network (DFDNet) enhances fine-grained out-of-distribution (OOD) detection by disentangling features, achieving significant performance improvements across multiple datasets.
Quick Take
The Dual Feature Decoupling Network (DFDNet) enhances fine-grained out-of-distribution (OOD) detection by disentangling features, achieving significant performance improvements across multiple datasets. It includes a spatial-frequency decoupling module to preserve discriminative content and a reconstruction-guided module to eliminate non-discriminative information.
Key Points
- DFDNet addresses challenges in fine-grained OOD detection with feature disentanglement.
- Spatial-frequency decoupling preserves classification-relevant content while suppressing style information.
- Reconstruction-guided module enhances high-level semantic representations at the pixel level.
- Extensive experiments show competitive performance across various datasets.
Article Content
From source RSS / original summaryarXiv:2606. 05536v1 Announce Type: new Abstract: Out-of-distribution detection (OOD) is an indispensable technique when applying machine learning models to real-world scenarios. Most existing OOD detection methods have been developed under the idealized assumption of large inter-class distributional differences, while largely overlooking fine-grained tasks characterized by subtle variations, such as medical image classification and vehicle recognition.
The high visual similarity among fine-grained subcategories, together with the interference of background factors, makes OOD detection extremely challenging. To tackle this problem, we propose a novel Dual Feature Decoupling Network (DFDNet), which addresses fine-grained OOD detection from the perspective of feature disentanglement. The proposed DFDNet comprises two key components: a spatial-frequency decoupling module and a reconstruction-guided decoupling module.
The spatial-frequency decoupling module is designed to preserve content features that are discriminative for classification while suppressing task-irrelevant style information. On the other hand, the reconstruction-guided decoupling module introduces a novel pixel-level adversarial reconstruction task to further remove low-level, non-discriminative information and enhance category-specific high-level semantic representations.
Extensive experiments demonstrate that our method achieves competitive performance improvements on multiple datasets.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.