摘要
近年来基于深度学习的细粒度分类是研究的热点,细粒度分类的主要方法是先找出分类对象再分类。找出分类对象的方法中主要分为两种:强监督与弱监督,强监督需要使用昂贵的人工标签,为了减少人工标注成本,提出一种基于FCN的图像感兴趣区域的分割与提取,并利用分割的图像进一步训练网络提高正确率。
In recent years, fine-grained classification based on deep learning is a hot topic. The main method of fine-grained classification is to first find the classification object and then classify it. There are two main methods for finding classification objects: strong supervision and weak supervision. Strong supervision requires the use of expensive manual labels. In order to reduce the cost of manual labeling, mainly proposes a segmentation and extraction of image regions based on improved FCN. The method and use the segmented image to further train the network to improve the correct rate.
引文
[1]P. Welinder,S. Branson,T. Mita,C. Wah,F. Schroff,S. Belongie,P. Perona. Caltech-UCSD Birds 200,2010.
[2]A. Khosla,N. Jayadevaprakash,B. Yao,F.-F. Li. Dataset for Fine-Grained Image Categorization. In First Workshop on Fine-Grained Visual Categorization,CVPR,2011.
[3]M.-E. Nilsback,A. Zisserman. Automated Flower Classification over a Large Number of Classes. In Proceedings of the Indian Conference on Computer Vision. Graphics and Image Processing,2008.
[4]S. Maji,J. Kannala,E. Rahtu,M. Blaschko,A. Vedaldi.Fine-Grained Visual Classification of Aircraft. Technical Report,2013.
[5]T. Xiao,Y. Xu,K. Yang,J. Zhang,Y. Peng,Z. Zhang. The Application of Two-Level Attention Models in Deep Convolutional Neural Network for Fine-Grained Image Classification. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,Boston,MA,Jun. 2015:842-850.
[6]M. Simon,E. Rodner. Neural Activation Constellations:Unsupervised Part Model Discovery with Convolutional Networks. In Proceedings of IEEE International Conference on Computer Vision,Sandiago,Chile,Dec. 2015:1143-1151.
[7]T.Y. Lin,A. RoyChowdhury,S. Maji. Bilinear CNN Models for Fine-Grained Visual Recognition. In Proceedings of IEEE International Conference on Computer Vision,Sandiago,Chile,Dec. 2015:1449-1457.
[8]Shelhamer E,Long J,Darrell T. Fully Convolutional Networks for Semantic Segmentation[M]. IEEE Computer Society,2017.
[9]Rahman M A, Wang Y. Optimizing Intersection-Over-Union in Deep Neural Networks for Image Segmentation[M]. Advances in Visual Computing. Springer International Publishing,2016:234-244.
[10]X. Liu,T. Xia,J. Wang,Y. Q. Lin. Fully Convolutional Attention Localization Networks:Efficient Attention Localization for Fine-Grained Recognition. arXiv:1603.06765,2016.
[11]N. Zhang,J. Donahue,R. Girshick,T. Darrell. Part-Based R-CNNs for Fine-Grained Category Detection. In Proceedings of the 13th European Conference on Computer Vision,Springer,Zurich,Switzerland,2014,8689:834-849.
[12]D. Lin,X. Y. Shen,C. W. Lu,J. Y. Jia. Deep LAC:Deep Localization, Alignment and Classification for Fine-Grained Recognition. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,IEEE,Boston,USA,2015:1666-1674.