Autonomous Multi-Robot Visual Monitoring
The project addresses fundamental research to develop novel autonomous and adaptive monitoring systems for natural resources across large spatio-temporal scales using networks of aerial robots equipped with visual sensors. The robots will be able to autonomously adapt to the observed phenomena and to multiple, often conflicting, time-varying constraints and mission specifications, greatly improving the precision of the collected data, and allowing spatio-temporal scalability. This is achieved by considering tradeoffs between visual sensing, real-time trajectory planning, decision making, and system optimization.
Acknowledgments: NSF Grant 1724341 from Smart and Autonomous Systems program
Weakly Supervised Visual Learning
The goal is to adapt models learned in one environment to another new environment with little to no additional supervision.
-
Sample Publications
-
Domain Adaptive Semantic Segmentation Using Weak Labels
S. Paul, Y. Tsai, S. Schulter, A. Roy-Chowdhury, M. Chandraker, European Conference on Computer Vision (ECCV), 2020.
-
Camera On-boarding for Person Re-identification using Hypothesis Transfer Learning
Sk. M. Ahmed*, A. R. lejbolle*, R. Panda, A. Roy-Chowdhury, IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2020. (*joint first authors)
-
Learning from Trajectories via Subgoal Discovery
S. Paul, Jeroen van Baar, A. Roy-Chowdhury, Advances in Neural Information Processing Systems (NeurIPS), 2019
-
Exploiting Transitivity for Learning Person Re-identification Models on a Budget
Roy, S. Paul, N.E. Young, A. Roy-Chowdhury, Computer Vision and Pattern Recognition (CVPR), 2018
-
FFNet: Video Fast-Forwarding via Reinforcement Learning
S. Lan, R. Panda, Q. Zhu, and A. Roy-Chowdhury, CVPR, 2018.
-
Contemplating Visual Emotions: Understanding and Overcoming Dataset Bias
R. Panda, J. Zhang, H. Li, J. Y., Lee, X. Lu, and A. Roy-Chowdhury, European Conference on Computer Vision, 2018.
-
W-TALC: Weakly Supervised Temporal Activity Localization and Classification
S. Paul, S. Roy, and A. Roy-Chowdhury, European Conference on Computer Vision, 2018.
-
Multi-robot Coordination and Decision Making
This part of the project will focus on multi-robot information gathering, decision making and field testing and evaluation. More information can be found here.