J
Jia Deng
Researcher at Princeton University
Publications - 158
Citations - 110718
Jia Deng is an academic researcher from Princeton University. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 50, co-authored 148 publications receiving 73461 citations. Previous affiliations of Jia Deng include University of Michigan & Carnegie Mellon University.
Papers
More filters
Posted Content
Non-deep Networks
TL;DR: In this paper, the authors use parallel subnetworks instead of stacking one layer after another, which helps effectively reduce depth while maintaining high performance and achieves state-of-the-art performance.
Posted Content
DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras
Zachary Teed,Jia Deng +1 more
TL;DR: DROID-SLAM as mentioned in this paper is a deep learning-based SLAM system that consists of recurrent iterative updates of camera pose and pixelwise depth through a Dense Bundle Adjustment layer.
Proceedings Article
Dynamically grown generative adversarial networks
TL;DR: In this article, the authors propose a method to dynamically grow a GAN during training, optimizing the network architecture and its parameters together with automation, which enjoys the benefits of both eased training because of progressive growing and improved performance because of broader architecture design space.
Posted Content
Compositional Temporal Visual Grounding of Natural Language Event Descriptions.
TL;DR: This work develops a unified deep architecture, CTG-Net, to perform temporal grounding of natural language event descriptions to videos, and demonstrates that this system outperforms prior state-of-the-art methods on the DiDeMo, Tempo-TL, and TemPO-HL temporal grounding datasets.
Posted Content
RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching.
TL;DR: In this paper, a new deep architecture for rectified stereo based on the optical flow network RAFT is introduced, which more efficiently propagates information across the image and ranks first on the Middlebury leaderboard, outperforming the next best method on 1px error.