H
Hsu Chih-Wei
Researcher at MediaTek
Publications - 183
Citations - 17444
Hsu Chih-Wei is an academic researcher from MediaTek. The author has contributed to research in topics: Motion vector & Encoder. The author has an hindex of 27, co-authored 183 publications receiving 16628 citations. Previous affiliations of Hsu Chih-Wei include National Defense Medical Center & National Taiwan University.
Papers
More filters
A Practical Guide to Support Vector Classication
TL;DR: A simple procedure is proposed, which usually gives reasonable results and is suitable for beginners who are not familiar with SVM.
Journal ArticleDOI
A comparison of methods for multiclass support vector machines
Hsu Chih-Wei,Chih-Jen Lin +1 more
TL;DR: Decomposition implementations for two "all-together" multiclass SVM methods are given and it is shown that for large problems methods by considering all data at once in general need fewer support vectors.
A Comparison of Methods for Multi-class Support Vector Machines
Hsu Chih-Wei,Chih-Jen Lin +1 more
TL;DR: These experiments indicate that the “one-against-one” and DAG methods are more suitable for practical use than the other methods, and show that for large problems methods by considering all data at once in general need fewer support vectors.
Journal ArticleDOI
Sample Adaptive Offset in the HEVC Standard
Chih-Ming Fu,Elena Alshina,Alexander Alshin,Yu-Wen Huang,Chen Ching-Yeh,Chia-Yang Tsai,Hsu Chih-Wei,Shaw-Min Lei,Jeong-Hoon Park,Woo-Jin Han +9 more
TL;DR: This paper provides a technical overview of a newly added in-loop filtering technique, sample adaptive offset (SAO), in High Efficiency Video Coding (HEVC), to reduce sample distortion by first classifying reconstructed samples into different categories, obtaining an offset for each category, and then adding the offset to each sample of the category.
Proceedings ArticleDOI
Discriminative Frequent Pattern Analysis for Effective Classification
TL;DR: This paper develops a strategy to set minimum support in frequent pattern mining for generating useful patterns, and demonstrates that the frequent pattern-based classification framework can achieve good scalability and high accuracy in classifying large datasets.