G
Gang Sun
Researcher at Chinese Academy of Sciences
Publications - 14
Citations - 21176
Gang Sun is an academic researcher from Chinese Academy of Sciences. The author has contributed to research in topics: Convolutional neural network & Feature (computer vision). The author has an hindex of 7, co-authored 12 publications receiving 8592 citations. Previous affiliations of Gang Sun include SenseTime.
Papers
More filters
Journal ArticleDOI
Multi-Level Discriminative Dictionary Learning With Application to Large Scale Image Classification
TL;DR: This paper proposes a novel multi-level discriminative dictionary learning method that achieves excellent accuracy and competitive computation cost compared with other sparse coding methods for large scale image classification.
Proceedings ArticleDOI
Multi-level Discriminative Dictionary Learning towards Hierarchical Visual Categorization
TL;DR: This paper proposes a novel dictionary learning method by taking advantage of hierarchical category correlation, where each internode of the hierarchical category structure is learnt for visual categorization, and the dictionaries in different layers are learnt to exploit the discriminative visual properties of different granularity.
Posted Content
Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks
TL;DR: Gathering and Excite as mentioned in this paper proposes a pair of operators: gather and excite, which redistributes the pooled information to local features, which can be integrated directly in existing architectures to improve their performance.
Proceedings ArticleDOI
Adaptive multi-task learning for fine-grained categorization
TL;DR: This paper proposes a novel multi-task learning method to adaptively share information that captures the relationships among tasks and identifies the disparities of each task simultaneously, thus can flexibly exploit the shared information.
Journal ArticleDOI
Accurate and efficient cross-domain visual matching leveraging multiple feature representations
TL;DR: To integrate the discriminative power of multiple features, a data-driven, query specific feature fusion model is developed, which estimates the relative importance of the individual feature dimensions as well as the weight vector among multiple features simultaneously.