Learning a Discriminative Filter Bank within a CNN for Fine-grained Recognition

CVPR 2018  ·  Yaming Wang, Vlad I. Morariu, Larry S. Davis ·

Compared to earlier multistage frameworks using CNN features, recent end-to-end deep approaches for fine-grained recognition essentially enhance the mid-level learning capability of CNNs. Previous approaches achieve this by introducing an auxiliary network to infuse localization information into the main classification network, or a sophisticated feature encoding method to capture higher order feature statistics. We show that mid-level representation learning can be enhanced within the CNN framework, by learning a bank of convolutional filters that capture class-specific discriminative patches without extra part or bounding box annotations. Such a filter bank is well structured, properly initialized and discriminatively learned through a novel asymmetric multi-stream architecture with convolutional filter supervision and a non-random layer initialization. Experimental results show that our approach achieves state-of-the-art on three publicly available fine-grained recognition datasets (CUB-200-2011, Stanford Cars and FGVC-Aircraft). Ablation studies and visualizations are provided to understand our approach.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Fine-Grained Image Classification CUB-200-2011 DFB Accuracy 87.4% # 55
Fine-Grained Image Classification CUB-200-2011 DFL-CNN Accuracy 87.4 # 20
Fine-Grained Image Classification FGVC Aircraft DFB-CNN Accuracy 92.0% # 41
Fine-Grained Image Classification Stanford Cars DFL-CNN Accuracy 93.8% # 52

Methods


No methods listed for this paper. Add relevant methods here