H
Honghui Shi
Researcher at University of Oregon
Publications - 65
Citations - 3681
Honghui Shi is an academic researcher from University of Oregon. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 20, co-authored 32 publications receiving 2007 citations. Previous affiliations of Honghui Shi include IBM & University of Illinois at Urbana–Champaign.
Papers
More filters
Proceedings ArticleDOI
Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi-Supervised Semantic Segmentation
TL;DR: In this article, a generic classification network equipped with convolutional blocks of different dilated rates was designed to produce dense and reliable object localization maps and effectively benefit both weakly and semi-supervised semantic segmentation.
Proceedings ArticleDOI
HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation
TL;DR: HigherHRNet is presented, a novel bottom-up human pose estimation method for learning scale-aware representations using high-resolution feature pyramids that surpasses all top-down methods on CrowdPose test and achieves new state-of-the-art result on COCO test-dev, suggesting its robustness in crowded scene.
Proceedings ArticleDOI
Self-Similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-Identification
TL;DR: A Self-similarity Grouping (SSG) approach, which exploits the potential similarity of unlabeled samples to build multiple clusters from different views automatically, and introduces a clustering-guided semisupervised approach named SSG ++ to conduct the one-shot domain adaption in an open set setting.
Journal ArticleDOI
Horizontal Pyramid Matching for Person Re-Identification
Yang Fu,Yunchao Wei,Yuqian Zhou,Honghui Shi,Gao Huang,Xinchao Wang,Zhiqiang Yao,Thomas S. Huang +7 more
TL;DR: A simple yet effective Horizontal Pyramid Matching (HPM) approach to fully exploit various partial information of a given person, so that correct person candidates can be still identified even even some key parts are missing.
Posted Content
Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation
TL;DR: It is found that varying dilation rates can effectively enlarge the receptive fields of convolutional kernels and more importantly transfer the surrounding discriminative information to non-discriminative object regions, promoting the emergence of these regions in the object localization maps.