2017 4th IAPR Asian Conference on Pattern Recognition (ACPR)最新文献_第8页

PCA-LDANet: A Simple Feature Learning Method for Image Classification PCA-LDANet:一种简单的图像分类特征学习方法

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.36

Yukun Ge, Jiani Hu, Weihong Deng

{"title":"PCA-LDANet: A Simple Feature Learning Method for Image Classification","authors":"Yukun Ge, Jiani Hu, Weihong Deng","doi":"10.1109/ACPR.2017.36","DOIUrl":"https://doi.org/10.1109/ACPR.2017.36","url":null,"abstract":"In this paper, we propose a simple and effective feature learning architecture for image classification that is based on very basic data processing components: 1) principal component analysis (PCA); 2) linear discriminant analysis (LDA); and 3) binary hashing and blockwise histograms. In this architecture, the PCA is employed to reconstruct patches of input images, and the LDA is employed to learn filter banks. This is followed by simple binary hashing and blockwise histograms for indexing. This architecture is motivated by LDANet and PCANet, thus called the PCA LDA Network (PCA-LDANet). They have some similarities in their topologies. We have tested the PCA-LDANet on two visual datasets for different tasks, including the Facial Recognition Technology (FERET) dataset for face recognition; and MNIST dataset for hand-written digit recognition. To explore the properties and essence of these architectures, we just conduct experiments on the one-stage networks. It is enough to explain the issue properly. Experimental results show that the PCA-LDANet-1 outperforms both PCANet-1 and LDANet-1 on both datasets. The experimental results demonstrate the effectiveness and distinctiveness of the PCA-LDANet; and the important role of PCA patch reconstruction in the PCA-LDANet.","PeriodicalId":426561,"journal":{"name":"2017 4th IAPR Asian Conference on Pattern Recognition (ACPR)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128751931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

Learning Principal Orientations Descriptor for Action Recognition 动作识别的主取向描述符学习

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.28

Lei Chen, Jiwen Lu, Zhanjie Song, Jie Zhou

引用次数: 1

Travel Time-Dependent Maximum Entropy Inverse Reinforcement Learning for Seabird Trajectory Prediction 基于旅行时变最大熵逆强化学习的海鸟轨迹预测

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.20

Tsubasa Hirakawa, Takayoshi Yamashita, K. Yoda, Toru Tamaki, H. Fujiyoshi

引用次数: 3

Deep Feature Similarity for Generative Adversarial Networks 生成对抗网络的深度特征相似度

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.47

Xianxu Hou, Ke Sun, G. Qiu

引用次数: 3

Motion Vector Based Data Association for On-Line Multi-object Tracking 基于运动矢量的在线多目标跟踪数据关联

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.54

Cong Ma, Z. Miao, Xiao-Ping Zhang, Min Li

引用次数: 0

Discriminative Transfer Learning Siamese CNN for Person Re-identification 鉴别迁移学习暹罗CNN人再识别

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.119

Yuan Tian, Cairong Zhao, Kang Chen, Yipeng Chen, Zhihua Wei, D. Miao

引用次数: 0

Re-ranking Person Re-identification with Local Discriminative Information 基于局部判别信息的人物重排序再识别

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.1

Kezhou Chen, N. Sang, Zhiqiang Li, Changxin Gao, Ruolin Wang

{"title":"Re-ranking Person Re-identification with Local Discriminative Information","authors":"Kezhou Chen, N. Sang, Zhiqiang Li, Changxin Gao, Ruolin Wang","doi":"10.1109/ACPR.2017.1","DOIUrl":"https://doi.org/10.1109/ACPR.2017.1","url":null,"abstract":"Most existing metric learning based person reidentification methods try to learn a global distance metric to measure the similarity between person images. But owing to the large intra-class variations, pedestrian data follows very irregular distribution in the feature space. The global metric model can hardly exploit the discriminative information from local distribution. Thus, due to the higher similarity of distribution, local information should be elaborately mined and exploited to improve the matching accuracy, especially for some hard positive images. In this paper, we propose to combine the global metric and local information to resolve failure matching cases. Detailly, for a testing pair, positive pairs in the training set whose feature differences are similar with given testing pair under global metric are firstly searched. If most of these positive pairs are located in the local range of the testing pair, the global metric is thus believed to reflect the similarity relationship in this local area. According to the degree of local discriminative information being represented in global metric, testing pair is derived based on the global metric as well as the given pair's local information. Finally, all gallery images are re-ranked according to the combined similarity scores. Experimental results on VIPeR, PRID450S and Market-1501 datasets clearly demonstrate the effectiveness of the proposed method.","PeriodicalId":426561,"journal":{"name":"2017 4th IAPR Asian Conference on Pattern Recognition (ACPR)","volume":"448 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114096855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Efficient Speaker Naming via Deep Audio-Face Fusion and End-to-End Attention Model 基于深度音面融合和端到端注意模型的高效说话人命名

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.13

Xin Liu, Jiajia Geng, Haibin Ling

{"title":"Efficient Speaker Naming via Deep Audio-Face Fusion and End-to-End Attention Model","authors":"Xin Liu, Jiajia Geng, Haibin Ling","doi":"10.1109/ACPR.2017.13","DOIUrl":"https://doi.org/10.1109/ACPR.2017.13","url":null,"abstract":"Speaker naming has recently received wide attention in identifying the speaking character in a movie video, and it is an extremely challenging topic mainly attributed to the significant variation of facial appearance. Motivated by multimodal applications, we present an efficient speaker naming approach via deep audio-face fusion and end-to-end attention model. First, we start with LSTM-encoding of acoustic feature and VGG-encoding of face images, and then exploit an end-to-end common attention vector by convolution-softmax encoding of their locally concatenated features, whereby the face attention vector can be well discriminated. Further, we apply the low-rank bilinear model to efficiently fuse the face attention vector and acoustic feature vector, whereby the joint audio-face representation can be discriminatively obtained for speaker naming. In addition, we address another acoustic feature representation scheme by convolution-encoding, which can replace LSTM in networks to speed up the training process. The experimental results have shown that our proposed speaker naming approach yields comparative and even better results than the state-of-the-art counterparts.","PeriodicalId":426561,"journal":{"name":"2017 4th IAPR Asian Conference on Pattern Recognition (ACPR)","volume":"189 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124188886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Learning a Smart Convolutional Neural Network with High-Level Semantic Information 基于高级语义信息的智能卷积神经网络学习

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.87

Xinshu Qiao, Chunyan Xu, Jian Yang, Jiatao Jiang

引用次数: 1

A New GVF Arrow Pattern for Character Segmentation from Double Line License Plate Images 一种新的GVF箭头模式用于双线车牌图像的字符分割

2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) Pub Date : 2017-11-01 DOI: 10.1109/ACPR.2017.45

P. Shivakumara, Aishik Konwer, A. Bhowmick, Vijeta Khare, U. Pal, Tong Lu

{"title":"A New GVF Arrow Pattern for Character Segmentation from Double Line License Plate Images","authors":"P. Shivakumara, Aishik Konwer, A. Bhowmick, Vijeta Khare, U. Pal, Tong Lu","doi":"10.1109/ACPR.2017.45","DOIUrl":"https://doi.org/10.1109/ACPR.2017.45","url":null,"abstract":"License plate recognition is a live problem for several developing countries because of its many challenges. One of such challenges is character segmentation from double lines (alphabets in one line and numerals on another line) license plate images, where we can see touching between adjacent characters (horizontally) and lines (vertically). This is the major cause for poor recognition performance. Therefore, we propose a novel technique based on Gradient Vector Flow (GVF) to segment characters from double line license plate images. The proposed technique explores a new GVF arrow pattern, which represents spaces between lines and characters based on the fact that the force in concavity created between characters and lines according to the fact that curved shaped characters attract GVF arrows in unique fashion. This observation leads to find seed space patches for segmentation. The spatial coordinates of seed space patches are passed through Hough transform to find line separators. Next, the proposed technique searches for seed space patches, which are perpendicular to line separators to find character separators. Experimental results on double line license plate images show that the proposed technique is robust to touching, rotations, scaling, distortion, and outperforms the existing character segmentation methods. The recognition experiments before and after segmentation show that the proposed segmentation is significant in improving license plate recognition rate.","PeriodicalId":426561,"journal":{"name":"2017 4th IAPR Asian Conference on Pattern Recognition (ACPR)","volume":"1 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123300918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3