Laboratory for Computational Sensing and Robotics Seminar: Jana Kosecka

Description

Dr. Jana Kosecka, a professor of Computer Science at George Mason University, will give a talk entitled "Visual Representations for Navigation and Object Discovery" for the Laboratory for Computational Sensing and Robotics.

Abstract:

Deliberate navigation in previously unseen environments and detection of novel objects instances are some of the key functionalities of intelligent agents engaged in fetch and delivery tasks. While data-driven deep-learning approaches fueled rapid progress in object category recognition and semantic segmentation by exploiting large amounts of labelled data, extending this learning paradigm to robotic setting comes with challenges. To overcome the need for large amount of labeled data for training object instance detectors, we use active self-supervision provided by a robot traversing an environment. The knowledge of ego-motion enables the agent to effectively associate multiple object hypotheses, which serve as training data for learning novel object embeddings from unlabeled data. The object detectors trained in this manner achieve higher mAP compared to off-the-shelf detectors trained on this limited data. I will describe an approach towards semantic target driven navigation, which entails finding a way through a complex environment to a target object. The proposed approach learns navigation policies on top of representations that capture spatial layout and semantic contextual cues. The choice of this representation exploits models trained on large standard vision datasets, enables better generalization and joint use of simulated environments and real images for effective training of navigation policies.