Y
Yulan Guo
Researcher at National University of Defense Technology
Publications - 196
Citations - 9683
Yulan Guo is an academic researcher from National University of Defense Technology. The author has contributed to research in topics: Computer science & Point cloud. The author has an hindex of 30, co-authored 164 publications receiving 5012 citations. Previous affiliations of Yulan Guo include Chinese Academy of Sciences & University of Western Australia.
Papers
More filters
Journal ArticleDOI
InStereo2K: a large real dataset for stereo matching in indoor scenes
TL;DR: A new stereo dataset called InStereo2K is introduced, which contains 2050 pairs of stereo images with highly accurate groundtruth disparity maps and can significantly improve the performance of several latest networks on the Middlebury 2014 dataset.
Journal ArticleDOI
Deep point-to-subspace metric learning for sketch-based 3D shape retrieval
TL;DR: This work proposes a Deep Point-to-Subspace Metric Learning (DPSML) framework to project a sketch into a feature vector and a 3D shape into a subspace spanned by a few selected basis feature vectors to reduce the redundancy of 3D shapes and introduces a Representative-View Selection (RVS) module to select the most representative views of a3D shape.
Proceedings Article
Axiom−based Grad−CAM: Towards Accurate Visualization and Explanation of CNNs
TL;DR: XGrad-CAM as discussed by the authors is an axiom-based version of the gradient-cAM, which is able to achieve better visualization performance than the original gradientcAM.
Proceedings ArticleDOI
Semantic-Aware Domain Generalized Segmentation
TL;DR: This paper addresses domain generalized semantic segmentation, where a segmentation model is trained to be domain-invariant without using any target domain data, and proposes a framework including two novel modules: Semantic-Aware Normalization (SAN) and Semantic -Aware Whitening (SAW).
Journal ArticleDOI
Scale space clustering evolution for salient region detection on 3D deformable shapes
TL;DR: A novel framework that relies on clustering of a data set derived from the scale space of the auto diffusion function, which consists of three major techniques: scalar field construction, shape segmentation initialization and salient region detection.