Y
Yue Zhao
Researcher at The Chinese University of Hong Kong
Publications - 24
Citations - 1606
Yue Zhao is an academic researcher from The Chinese University of Hong Kong. The author has contributed to research in topics: Computer science & Action (philosophy). The author has an hindex of 11, co-authored 20 publications receiving 1092 citations. Previous affiliations of Yue Zhao include Tsinghua University & SenseTime.
Papers
More filters
Proceedings ArticleDOI
Temporal Action Detection with Structured Segment Networks
TL;DR: In this article, a structured segment network (SSN) is proposed to model the temporal structure of each action instance via a structured temporal pyramid, and a decomposed discriminative model comprising two classifiers, respectively for classifying actions and determining completeness.
Proceedings ArticleDOI
FineGym: A Hierarchical Video Dataset for Fine-Grained Action Understanding
TL;DR: FineGym is a new dataset built on top of gymnasium videos that provides temporal annotations at both action and sub-action levels with a three-level semantic hierarchy and systematically investigates different methods on this dataset and obtains a number of interesting findings.
Journal ArticleDOI
Temporal Action Detection with Structured Segment Networks
TL;DR: This paper addresses an important and challenging task, namely detecting the temporal intervals of actions in untrimmed videos by presenting a framework called structured segment network (SSN), built on temporal proposals of actions.
Posted Content
A Pursuit of Temporal Accuracy in General Activity Detection
TL;DR: A generic framework that can accurately detect a wide variety of activities from untrimmed videos is proposed that significantly outperforms the existing state-of-the-art methods, demonstrating superior accuracy and strong adaptivity in handling activities with various temporal structures.
Proceedings Article
Trajectory Convolution for Action Recognition
Yue Zhao,Yuanjun Xiong,Dahua Lin +2 more
TL;DR: This work proposes a new CNN architecture TrajectoryNet, which incorporates trajectory convolution, a new operation for integrating features along the temporal dimension, to replace the existing temporal convolution.