Mengyuan Liu

TL;DR: A Deformable Pose Traversal Convolution Network is proposed that applies one-dimensional convolution to traverse the 3D pose for its representation and optimizes the convolution kernel for each joint, by considering contextual joints with various weights.

TL;DR: A local point detector is developed by sampling local points based on both motion and shape clues to represent human activities in depth sequences and is robust to partial occlusions in depth data, and also robust to the changes of pose, illumination and background to some extent.

TL;DR: A traditional classification algorithm based on digital image processing was attempted, and a defect classification algorithms based on convolutional neural network was proposed, which can achieve a fairly high classification accuracy (95.7%), which is much higher than the traditional method, and the new method has stronger stability than thetraditional one.

TL;DR: An effective depth feature representation is developed based on the fusion of 2D and 3D auto-correlation of gradients features based on a weighted fusion strategy to assign different weights to the classifier probability outputs associated with different features, thereby providing more flexibility in the final decision making.

TL;DR: A local spatio-temporal descriptor for action recognistion from depth video sequences, which is capable of distinguishing similar actions as well as coping with different speeds of actions is presented.