Y
Yawei Luo
Researcher at University of Technology, Sydney
Publications - 20
Citations - 1479
Yawei Luo is an academic researcher from University of Technology, Sydney. The author has contributed to research in topics: Computer science & Domain (software engineering). The author has an hindex of 11, co-authored 15 publications receiving 861 citations. Previous affiliations of Yawei Luo include Huazhong University of Science and Technology.
Papers
More filters
Proceedings ArticleDOI
Taking a Closer Look at Domain Shift: Category-Level Adversaries for Semantics Consistent Domain Adaptation
TL;DR: In this paper, a category-level adversarial network is proposed to enforce local semantic consistency during the trend of global alignment, where the weight of the adversarial loss for well aligned features is reduced, while increasing the strength of adversarial force for poorly aligned features.
Posted Content
Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation
TL;DR: A category-level adversarial network is introduced, aiming to enforce local semantic consistency during the trend of global alignment, to take a close look at the category- level data distribution and align each class with an adaptive adversarial loss.
Proceedings ArticleDOI
Significance-Aware Information Bottleneck for Domain Adaptive Semantic Segmentation
TL;DR: Zhang et al. as mentioned in this paper proposed a significance-aware information bottleneck (SIBAN), which enables a significanceaware feature purification before adversarial adaptation, which eases the feature alignment and stabilizes the adversarial training course.
Proceedings ArticleDOI
P-MVSNet: Learning Patch-Wise Matching Confidence Aggregation for Multi-View Stereo
TL;DR: This paper proposes a new end-to-end deep learning network of P-MVSNet for multi-view stereo based on isotropic and anisotropic 3D convolutions and achieves the state-of-the-art performance over many existing methods on multi-View stereo.
Book ChapterDOI
Macro-Micro Adversarial Network for Human Parsing
TL;DR: Compared with traditional adversarial networks, the proposed Macro-Micro Adversarial Net not only enforces local and semantic consistency explicitly, but also avoids the poor convergence problem of adversarial Networks when handling high resolution images.