K
Kaihua Zhang
Researcher at Nanjing University of Information Science and Technology
Publications - 98
Citations - 7681
Kaihua Zhang is an academic researcher from Nanjing University of Information Science and Technology. The author has contributed to research in topics: Video tracking & Feature (computer vision). The author has an hindex of 28, co-authored 91 publications receiving 6787 citations. Previous affiliations of Kaihua Zhang include Nanjing University & Hong Kong Polytechnic University.
Papers
More filters
Book ChapterDOI
Real-time compressive tracking
TL;DR: A simple yet effective and efficient tracking algorithm with an appearance model based on features extracted from the multi-scale image feature space with data-independent basis that performs favorably against state-of-the-art algorithms on challenging sequences in terms of efficiency, accuracy and robustness.
Journal ArticleDOI
Active contours with selective local or global segmentation: A new formulation and level set method
TL;DR: A novel region-based active contour model (ACM) with SBGFRLS has the property of selective local or global segmentation, which is more efficient to construct than the widely used signed distance function (SDF).
Book ChapterDOI
Fast Visual Tracking via Dense Spatio-temporal Context Learning
TL;DR: A novel explicit scale adaptation scheme is proposed, able to deal with target scale variations efficiently and effectively, and the Fast Fourier Transform is adopted for fast learning and detection in this work, which only needs 4 FFT operations.
Journal ArticleDOI
Active contours driven by local image fitting energy
TL;DR: A new region-based active contour model that embeds the image local information by introducing the local image fitting (LIF) energy to extract the localimage information is proposed and is able to segment images with intensity inhomogeneities.
Journal ArticleDOI
Fast Compressive Tracking
TL;DR: A simple yet effective and efficient tracking algorithm with an appearance model based on features extracted from a multiscale image feature space with dataindependent basis that performs favorably against state-of-the-art methods on challenging sequences in terms of efficiency, accuracy and robustness.