Learning a Discriminative Filter Bank Within a CNN for Fine-Grained Recognition
Authors:
- Yaming WangYWYaming Wang
- Vlad I. MorariuVMVlad I. Morariu
- Larry S. DavisLDLarry S. Davis
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 4148-4157
Abstract
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.Keywords
All Related Versions
This publication has 39 references indexed in Scilit:
- Going deeper with convolutions2015
- Hyper-class augmented and regularized deep learning for fine-grained image classification2015
- Fully convolutional networks for semantic segmentation2015
- Revisiting the Fisher vector for fine-grained classificationPattern Recognition Letters, 2014
- Similarity Comparisons for Interactive Fine-Grained Categorization2014
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Deformable Part Descriptors for Fine-Grained Recognition and Attribute Prediction2013
- Symbiotic Segmentation and Part Localization for Fine-Grained Categorization2013
- 3D Object Representations for Fine-Grained Categorization2013
- POOF: Part-Based One-vs.-One Features for Fine-Grained Categorization, Face Verification, and Attribute Estimation2013