Zhenjie Hou

  • Citations Per Year
Learn More
This paper presents a new framework for human action recognition from depth sequences. An effective depth feature representation is developed based on the fusion of 2D and 3D auto-correlation of gradients features. Specifically, depth motion maps (DMMs) are first employed to transform a depth sequence into three images capturing shape and motion cues. A(More)
Camera calibration is one of the key procedures for computer vision and the accuracy of camera calibration is of vital importance in terms of ensuring the systematic measurement accuracy. An automatic calibration method based on traditional camera calibration is proposed in this paper. First of all, calibrated image series can be gained by using improved(More)
The features of SIFT are widely used in image matching because of it’s invariantion to images scale and rotation, but 128-dimensional description of the feature point reduce the efficiency of the algorithm. This paper presents an improved algorithm of SIFT, using ring and sequence of each feature vector to ensure rotation invariance while reducing(More)
The temporal delay in human visual system results in latency between the occurrence of a visual stimulus and its perception by the observer, which is known as visual response latency. It has been an area of interest to biomedicine, telecine, and psychology over the years. This latency can be tested by transforming time into space, which can be easily(More)
Image matching is an important work in the field of computer image processing, the technology aligned same scene of two or more images obtained by different sensors in different or the same imaging conditions to determine the relationship between them. Based on the SURF algorithm, this paper adopts density threshold suppression strategies to reduce number(More)