Y
Yi-Lei Chen
Researcher at National Tsing Hua University
Publications - Â 16
Citations - Â 1025
Yi-Lei Chen is an academic researcher from National Tsing Hua University. The author has contributed to research in topics: Image compression & Transform coding. The author has an hindex of 11, co-authored 16 publications receiving 837 citations.
Papers
More filters
Proceedings ArticleDOI
A Generalized Low-Rank Appearance Model for Spatio-temporally Correlated Rain Streaks
Yi-Lei Chen,Chiou-Ting Hsu +1 more
TL;DR: This work proposes and generalizes a low-rank model from matrix to tensor structure in order to capture the spatio-temporally correlated rain streaks and removes rain streaks from image/video in a unified way.
Journal ArticleDOI
Simultaneous Tensor Decomposition and Completion Using Factor Priors
TL;DR: This paper proposes a method called simultaneous tensor decomposition and completion (STDC) that combines a rank minimization technique with Tucker model decomposition, and uses factor priors, which are usually known a priori in real-world tensor objects, to characterize the underlying joint-manifold drawn from the model factors.
Journal ArticleDOI
Detecting Recompression of JPEG Images via Periodicity Analysis of Compression Artifacts for Tampering Detection
Yi-Lei Chen,Chiou-Ting Hsu +1 more
TL;DR: This paper designs a robust detection approach which is able to detect either block-aligned or misaligned recompression in JPEG images, and shows it outperforms existing methods.
Journal ArticleDOI
Single-Image Dehazing via Optimal Transmission Map Under Scene Priors
TL;DR: This paper advocates the significance of accurate transmission estimation and recast the problem as deriving the optimal transmission map directly from the haze model under two scene priors, and introduces theoretic and heuristic bounds of scene transmission to guide the optimum.
Journal ArticleDOI
Subspace Learning for Facial Age Estimation Via Pairwise Age Ranking
Yi-Lei Chen,Chiou-Ting Hsu +1 more
TL;DR: The results on the age estimation demonstrate that the method outperforms classic subspace learning approaches, and the semi-supervised learning successfully incorporates the age ranks from unlabeled data under different scales and sources of data set.