Three papers on different topics were accepted to ECCV 2020 The paper titled " Domain Adaptive Semantic Segmentation Using Weak Labels" proposes a unified framework for both unsupervised domain adaptation when using the estimated pseudo-weak labels and the novel scenario of weakly-supervised adaptation when using weak labels acquired from human annotators. This work bridges the...
Papers on Non-Adversarial Video Synthesis with Learned Priors and Camera On-boarding for Person Re-identification using Hypothesis Transfer Learning are accepted to CVPR 2020. 1. The paper, Non-Adversarial Video Synthesis with Learned Priors, in CVPR 2020 proposes an approach to generate videos from latent noise vectors, without any reference input frames by jointly optimizing the input...
Our recent paper, Learning from Trajectories via Subgoal Discovery, in NeurIPS 2019 proposes an approach to discover subgoals to solve complex tasks, from a limited number of expert demonstrations. Using these discovered subgoals leads to significantly faster robotic task learning with none to limited hand-designing of rewards. This work is done in collaboration between UCR...
New 1.2 Million NSF grant “III: Medium: Efficient Collaborative Perception over Controllable Agent Networks” was awarded to a team of 3 CSE Professors: Vagelis Christidis (PI), Vassilis Tsotras, and Vagelis Papalexakis and 2 ECE Professors: Amit Roy-Chowdhury and Konstantinos Karydis.
Amit Roy-Chowdhury, Konstantinos Karydis, and Salman Asif, electrical and computer engineering professors, along with Fabio Pasqualetti, a mechanical engineering professor, receive a $1.2M grant from the Office of Naval Research’s Science of Artificial Intelligence Program. Using recent advances in computer vision, machine learning, signal processing, robotics, and controls, researchers will explore AI’s capability of both...
Recent papers in CVPR, ACM MM, and ACM ICMR have analyzed the issue of jointly analyzing visual and textual data. The CVPR-19 paper focuses on video moment retrieval based on associated captions, the ACM MM-18 paper considers web supervision for retrieval in joint image-text databases with limited labeled data, while the ICMR paper considered cross-modal...
A recent paper in T-IP proposed a novel deep network architecture for identifying image forgeries and presented results on multiple state-of-the-art datasets. A paper in NDSS-19 proposed how an adversary could design perturbations in video classification systems. Hybrid LSTM and Encoder-Decoder Architecture for Detection of Image Forgeries, IEEE Trans. on Image Processing, 2019. Adversarial Perturbations...
Two papers have been accepted to IEEE Conf. on Computer Vision and Pattern Recognition, 2018. The details of the papers can be found here. Video Fast Forwarding via Reinforcement Learning, Exploiting Transitivity for Person Re-Identification Two papers have been accepted to the European Conf. on Computer Vision, 2018. The details of the papers can be...
