arXiv subjects

Zhaoyang Li

Publications and source records attributed to Zhaoyang Li.

57 records · Page 4Linked to original sources

arXiv2018

Theoretical method for generating regular spatiotemporal pulsed-beam with controlled transverse-spatiotemporal dispersion

Herein we theoretically report a method that generates a transverse-spatiotemporal dispersion (T-STD), which is distinct from previous spatial, temporal, and longitudinal-spatiotemporal optics dispersions. By modulating T-STD, two not yet reported spatiotemporally structured beams (STSBs), i.e., the honeycomb beam and the picket-fence beam, can be generated in the space-time domain. The generated STSBs have novel and tunable periodic distributions. T-STD, STSB and their inner physical relationship are analyzed and introduced. We believe that this method might open a path towards new optical beams for potential applications, such as ultrafast optical fabrication and detection.

physics.optics↗

arXiv2016

Action Recognition Based on Joint Trajectory Maps Using Convolutional Neural Networks

Recently, Convolutional Neural Networks (ConvNets) have shown promising performances in many computer vision tasks, especially image-based recognition. How to effectively use ConvNets for video-based recognition is still an open problem. In this paper, we propose a compact, effective yet simple method to encode spatio-temporal information carried in $3D$ skeleton sequences into multiple $2D$ images, referred to as Joint Trajectory Maps (JTM), and ConvNets are adopted to exploit the discriminative features for real-time human action recognition. The proposed method has been evaluated on three public benchmarks, i.e., MSRC-12 Kinect gesture dataset (MSRC-12), G3D dataset and UTD multimodal human action dataset (UTD-MHAD) and achieved the state-of-the-art results.

cs.CV↗

arXiv2016

Combining ConvNets with Hand-Crafted Features for Action Recognition Based on an HMM-SVM Classifier

This paper proposes a new framework for RGB-D-based action recognition that takes advantages of hand-designed features from skeleton data and deeply learned features from depth maps, and exploits effectively both the local and global temporal information. Specifically, depth and skeleton data are firstly augmented for deep learning and making the recognition insensitive to view variance. Secondly, depth sequences are segmented using the hand-crafted features based on skeleton joints motion histogram to exploit the local temporal information. All training se gments are clustered using an Infinite Gaussian Mixture Model (IGMM) through Bayesian estimation and labelled for training Convolutional Neural Networks (ConvNets) on the depth maps. Thus, a depth sequence can be reliably encoded into a sequence of segment labels. Finally, the sequence of labels is fed into a joint Hidden Markov Model and Support Vector Machine (HMM-SVM) classifier to explore the global temporal information for final recognition.

cs.CV↗