H
Hanxi Li
Researcher at Jiangxi Normal University
Publications - 70
Citations - 1486
Hanxi Li is an academic researcher from Jiangxi Normal University. The author has contributed to research in topics: Computer science & Deep learning. The author has an hindex of 15, co-authored 46 publications receiving 1353 citations. Previous affiliations of Hanxi Li include Beihang University & NICTA.
Papers
More filters
Proceedings ArticleDOI
Real-time visual tracking using compressive sensing
TL;DR: Real-time Com-pressive Sensing Tracking (RTCST) as mentioned in this paper exploits the signal recovery power of compressive sensing (CS), and adopts Dimensionality Reduction and a customized Orthogonal Matching Pursuit (OMP) algorithm to accelerate the CS tracking.
Journal ArticleDOI
DeepTrack: Learning Discriminative Feature Representations Online for Robust Visual Tracking
Hanxi Li,Yi Li,Fatih Porikli +2 more
TL;DR: This paper presents an efficient and very robust tracking algorithm using a single convolutional neural network for learning effective feature representations of the target object in a purely online manner and introduces a novel truncated structural loss function that maintains as many training samples as possible and reduces the risk of tracking error accumulation.
Proceedings ArticleDOI
DeepTrack: learning discriminative feature representations by convolutional neural networks for visual tracking
Hanxi Li,Yi Li,Fatih Porikli +2 more
TL;DR: A model-free tracker that outperforms the existing state-of-the-art algorithms and rarely loses the track of the target object is proposed and a class-specific version of the proposed method that is tailored for tracking of a particular object class such as human faces is introduced.
Journal ArticleDOI
On the Dual Formulation of Boosting Algorithms
Chunhua Shen,Hanxi Li +1 more
TL;DR: It is theoretically proved that approximately, ℓ1-norm-regularized AdaBoost maximizes the average margin, instead of the minimum margin, and the duality formulation also enables us to develop column-generation-based optimization algorithms, which are totally corrective.
Book ChapterDOI
Robust online visual tracking with a single convolutional neural network
TL;DR: This work introduces a novel truncated structural loss function that maintains as many training samples as possible and reduces the risk of tracking error accumulation, thus drift, by accommodating the uncertainty of the model output.