Y
Yao Xue
Researcher at Xi'an Jiaotong University
Publications - 22
Citations - 362
Yao Xue is an academic researcher from Xi'an Jiaotong University. The author has contributed to research in topics: Computer science & Pixel. The author has an hindex of 9, co-authored 16 publications receiving 268 citations. Previous affiliations of Yao Xue include University of Alberta.
Papers
More filters
Journal ArticleDOI
Scalable Mobile Image Retrieval by Exploring Contextual Saliency
Xiyu Yang,Xueming Qian,Yao Xue +2 more
TL;DR: The proposed mobile image retrieval approach first determines the relevant photos according to visual similarity, then mines salient features by exploring contextual saliency from multiple relevant images, and finally determines contributions of salient features for scalable retrieval.
Book ChapterDOI
Cell Counting by Regression Using Convolutional Neural Network
TL;DR: A supervised learning framework with Convolutional Neural Network is described and cast the cell counting task as a regression problem, where the global cell count is taken as the annotation to supervise training, instead of following the classification or detection framework.
Journal ArticleDOI
Food and Ingredient Joint Learning for Fine-Grained Recognition
TL;DR: This paper proposes an Attention Fusion Network (AFN) and Food-Ingredient Joint Learning module for fine-grained food and ingredients recognition and proposes a balance focal loss to optimize the feature expression ability of the network for ingredients.
Posted Content
Cell Detection in Microscopy Images with Deep Convolutional Neural Network and Compressed Sensing
Yao Xue,Nilanjan Ray +1 more
TL;DR: This paper proposes a convolutional neural network (CNN)-based cell detection method that uses encoding of the output pixel space to encode the output space to a compressed vector of fixed dimension and demonstrates the first successful use of CNN with CS-based output space encoding.
Journal ArticleDOI
Landmark Summarization With Diverse Viewpoints
TL;DR: An approach for summarizing a collection of landmark images from diverse viewpoints with content overlap by viewpoint album (VA) generation is presented and experimental results show the effectiveness of the proposed landmark summarization approach.