H
Hao Su
Researcher at University of California, San Diego
Publications - 364
Citations - 82843
Hao Su is an academic researcher from University of California, San Diego. The author has contributed to research in topics: Computer science & Point cloud. The author has an hindex of 57, co-authored 302 publications receiving 55902 citations. Previous affiliations of Hao Su include Philips & Jiangxi University of Science and Technology.
Papers
More filters
Journal ArticleDOI
Drying Affects the Fiber Network in Low Molecular Weight Hydrogels
Laura L. E. Mears,Emily R. Draper,Emily R. Draper,Ana M. Castilla,Hao Su,Zhuola,Bart Dietrich,Bart Dietrich,Michael C. Nolan,Michael C. Nolan,Gregory N. Smith,James Doutch,Sarah E. Rogers,Riaz Akhtar,Honggang Cui,Dave J. Adams,Dave J. Adams +16 more
TL;DR: It is shown that the assumption that drying does not affect the network is not always correct, and small angle neutron scattering (SANS) is used to probe low molecular weight hydrogels formed by the self-assembly of dipeptides.
Journal ArticleDOI
Fine-grained semi-supervised labeling of large shape collections
TL;DR: A multi-label semi-supervised approach that takes as input a large shape collection of a given category with associated sparse and noisy labels, and outputs cleaned and complete labels for each shape, which yields results that are superior to state-of-the-art semi- supervised learning techniques.
Journal ArticleDOI
Comfort-Centered Design of a Lightweight and Backdrivable Knee Exoskeleton
Junlin Wang,Xiao Li,Tzu-Hao Huang,Shuangyue Yu,Yanjun Li,Tianyao Chen,Alessandra Carriero,Mooyeon Oh-Park,Hao Su +8 more
TL;DR: Kinematic simulations demonstrate that misalignment between the robot joint and knee joint can be reduced by 74% at maximum knee flexion and a low impedance mechanical transmission reduces the reflected inertia and damping of the actuator to human, thus the exoskeleton is highlybackdrivable.
Journal ArticleDOI
Deep part induction from articulated object pairs
TL;DR: In this article, the authors explore how the observation of different articulation states provides evidence for part structure and motion of 3D objects and propose a neural network architecture with three modules that respectively propose correspondences, estimate 3D deformation flows, and perform segmentation.
Posted Content
Extending Adversarial Attacks and Defenses to Deep 3D Point Cloud Classifiers
Daniel F. Liu,Ronald Yu,Hao Su +2 more
TL;DR: Overall, it is found that 3D point cloud classifiers are weak to adversarial attacks, but they are also more easily defensible compared to 2D image classifiers.