skip to main content

Search for: All records

Creators/Authors contains: "Gao, Yue"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. null (Ed.)
    3D object detection is an important yet demanding task that heavily relies on difficult to obtain 3D annotations. To reduce the required amount of supervision, we propose 3DIoUMatch, a novel semi-supervised method for 3D object detection applicable to both indoor and outdoor scenes. We leverage a teacher-student mutual learning framework to propagate information from the labeled to the unlabeled train set in the form of pseudo-labels. However, due to the high task complexity, we observe that the pseudo-labels suffer from significant noise and are thus not directly usable. To that end, we introduce a confidence-based filtering mechanism, inspired by FixMatch. We set confidence thresholds based upon the predicted objectness and class probability to filter low-quality pseudo-labels. While effective, we observe that these two measures do not sufficiently capture localization quality. We therefore propose to use the estimated 3D IoU as a localization metric and set category-aware self-adjusted thresholds to filter poorly localized proposals. We adopt VoteNet as our backbone detector on indoor datasets while we use PV-RCNN on the autonomous driving dataset, KITTI. Our method consistently improves state-of-the-art methods on both ScanNet and SUN-RGBD benchmarks by significant margins under all label ratios (including fully labeled setting). For example, when training using only 10% labeled data on ScanNet, 3DIoUMatch achieves 7.7 absolute improvement on mAP@0.25 and 8.5 absolute improvement on mAP@0.5 upon the prior art. On KITTI, we are the first to demonstrate semi-supervised 3D object detection and our method surpasses a fully supervised baseline from 1.8% to 7.6% under different label ratio and categories. 
    more » « less
  2. null (Ed.)
  3. Abstract

    Lipid droplet (LD) in vegetative tissues has recently been implicated in environmental responses in plants, but its regulation and its function in stress tolerance are not well understood. Here, we identified aMembrane Occupation and Recognition Nexus 1(MORN1) gene as a contributor to natural variations of stress tolerance through genome‐wide association study inArabidopsis thaliana. Characterization of its loss‐of‐function mutant and natural variants revealed that theMORN1gene is a positive regulator of plant growth, disease resistance, cold tolerance, and heat tolerance. The MORN1 protein is associated with the Golgi and is also partly associated with LD. Protein truncations that disrupt these associations abolished the biological function of the MORN1 protein. Furthermore, theMORN1gene is a positive regulator of LD abundance, and its role in LD number regulation and stress tolerance is highly linked. Therefore, this study identifies MORN1 as a positive regulator of LD abundance and a contributor to natural variations of stress tolerance. It implicates a potential involvement of Golgi in LD biogenesis and strongly suggests a contribution of LD to diverse processes of plant growth and stress responses.

    more » « less
  4. Metallic anodes (lithium, sodium, and zinc) are attractive for rechargeable battery technologies but are plagued by an unfavorable metal–electrolyte interface that leads to nonuniform metal deposition and an unstable solid–electrolyte interphase (SEI). Here we report the use of electrochemically labile molecules to regulate the electrochemical interface and guide even lithium deposition and a stable SEI. The molecule, benzenesulfonyl fluoride, was bonded to the surface of a reduced graphene oxide aerogel. During metal deposition, this labile molecule not only generates a metal-coordinating benzenesulfonate anion that guides homogeneous metal deposition but also contributes lithium fluoride to the SEI to improve Li surface passivation. Consequently, high-efficiency lithium deposition with a low nucleation overpotential was achieved at a high current density of 6.0 mA cm−2. A Li|LiCoO2cell had a capacity retention of 85.3% after 400 cycles, and the cell also tolerated low-temperature (−10 °C) operation without additional capacity fading. This strategy was applied to sodium and zinc anodes as well.

    more » « less
    more » « less