activity
20172022
most citedEvaluating Protein Transfer Learning with TAPE

79 citations · 276 across the 17 of their papers we have counts for

collaborators
Showing cs.CVShow all

7 papers · 1 filter

cs.CV2020

Active Learning for Video Description With Cluster-Regularized Ensemble Ranking

David M. Chan, Sudheendra Vijayanarasimhan, David A. Ross +1

Automatic video captioning aims to train models to generate text descriptions for all segments in a video, however, the most effective approaches require large amounts of manual an…

cs.CV20191 cited

Grounding Human-to-Vehicle Advice for Self-driving Vehicles

Jinkyu Kim, Teruhisa Misu, Yi-Ting Chen +2

Recent success suggests that deep neural control networks are likely to be a key component of self-driving vehicles. These networks are trained on large datasets to imitate human a…

cs.CV20195 cited

Periphery-Fovea Multi-Resolution Driving Model guided by Human Attention

Ye Xia, Jinkyu Kim, John Canny +2

Inspired by human vision, we propose a new periphery-fovea multi-resolution driving model that predicts vehicle speed from dash camera videos. The peripheral vision module of the m…

cs.CV2018

Diagnostic Visualization for Deep Neural Networks Using Stochastic Gradient Langevin Dynamics

Biye Jiang, David M. Chan, Tianhao Zhang +1

The internal states of most deep neural networks are difficult to interpret, which makes diagnosis and debugging during training challenging. Activation maximization methods are wi…

cs.CV2018

Label and Sample: Efficient Training of Vehicle Object Detector from Sparsely Labeled Data

Xinlei Pan, Sung-Li Chiang, John Canny

Self-driving vehicle vision systems must deal with an extremely broad and challenging set of scenes. They can potentially exploit an enormous amount of training data collected from…

cs.CV2018

Textual Explanations for Self-Driving Vehicles

Jinkyu Kim, Anna Rohrbach, Trevor Darrell +2

Deep neural perception and control networks have become key components of self-driving vehicles. User acceptance is likely to benefit from easy-to-interpret textual explanations wh…