Z
Zilong Huang
Researcher at Huazhong University of Science and Technology
Publications - 43
Citations - 4044
Zilong Huang is an academic researcher from Huazhong University of Science and Technology. The author has contributed to research in topics: Segmentation & Computer science. The author has an hindex of 16, co-authored 35 publications receiving 1899 citations. Previous affiliations of Zilong Huang include Tencent & University of Illinois at Urbana–Champaign.
Papers
More filters
Proceedings ArticleDOI
CCNet: Criss-Cross Attention for Semantic Segmentation
TL;DR: CCNet as mentioned in this paper proposes a recurrent criss-cross attention module to harvest the contextual information of all the pixels on its crisscross path, and then takes a further recurrent operation to finally capture the full-image dependencies from all pixels.
Posted Content
CCNet: Criss-Cross Attention for Semantic Segmentation
TL;DR: This work proposes a Criss-Cross Network (CCNet) for obtaining contextual information in a more effective and efficient way and achieves the mIoU score of 81.4 and 45.22 on Cityscapes test set and ADE20K validation set, respectively, which are the new state-of-the-art results.
Proceedings ArticleDOI
Weakly-Supervised Semantic Segmentation Network with Deep Seeded Region Growing
TL;DR: This paper proposes to train a semantic segmentation network starting from the discriminative regions and progressively increase the pixel-level supervision using by seeded region growing, and obtains the state-of-the-art performance.
Journal ArticleDOI
CCNet: Criss-Cross Attention for Semantic Segmentation
TL;DR: A Criss-Cross Network (CCNet) is proposed for obtaining full-image contextual information in a very effective and efficient way and achieves the mIoU scores of 81.9%, 45.76% and 55.47% on the Cityscapes test set, the ADE20K validation set and the LIP validation set respectively.
Posted Content
Devil in the Details: Towards Accurate Single and Multiple Human Parsing
TL;DR: This paper identifies several useful properties, including feature resolution, global context information and edge details, and performs rigorous analyses to reveal how to leverage them to benefit the human parsing task, resulting in a simple yet effective Context Embedding with Edge Perceiving (CE2P) framework for single human parsing.