D
Dawei Sun
Researcher at University of Illinois at Urbana–Champaign
Publications - 29
Citations - 667
Dawei Sun is an academic researcher from University of Illinois at Urbana–Champaign. The author has contributed to research in topics: Computer science & Metric (mathematics). The author has an hindex of 7, co-authored 22 publications receiving 296 citations. Previous affiliations of Dawei Sun include Tsinghua University & Intel.
Papers
More filters
Proceedings ArticleDOI
Learning Two-View Correspondences and Geometry Using Order-Aware Network
Jiahui Zhang,Dawei Sun,Zixin Luo,Anbang Yao,Lei Zhou,Tianwei Shen,Yurong Chen,Hongen Liao,Long Quan +8 more
TL;DR: This paper proposes Order-Aware Network, which infers the probabilities of correspondences being inliers and regresses the relative pose encoded by the essential matrix, and is built hierarchically and comprises three novel operations.
Journal ArticleDOI
Enabling Deep Learning on IoT Devices
TL;DR: Two ways to successfully integrate deep learning with low-power IoT products are explored.
Posted Content
Learning Two-View Correspondences and Geometry Using Order-Aware Network
Jiahui Zhang,Dawei Sun,Zixin Luo,Anbang Yao,Lei Zhou,Tianwei Shen,Yurong Chen,Long Quan,Hongen Liao +8 more
TL;DR: In this article, the authors propose an Order-Aware network, which infers the probabilities of correspondences being inliers and regresses the relative pose encoded by the essential matrix.
Proceedings ArticleDOI
Deeply-Supervised Knowledge Synergy
TL;DR: Deeply-supervised knowledge synergy (DKS) as mentioned in this paper is a new method aiming to train CNNs with improved generalization ability for image classification tasks without introducing extra computational cost during inference.
Posted Content
Deeply-supervised Knowledge Synergy
TL;DR: Deeply-supervised Knowledge Synergy is proposed, a new method aiming to train CNNs with improved generalization ability for image classification tasks without introducing extra computational cost during inference, and a novel synergy loss, which considers pairwise knowledge matching among all supervision branches.