Loading Events

« All Events

  • This event has passed.

Spring 2017 GRASP Seminar: Abhinav Shrivastava, Carnegie Mellon University, “The Small and the Rare: the Twin Menace of Visual Recognition”

February 24, 2017 @ 11:00 am - 12:00 pm

ABSTRACT

Over the past decade, we have seen tremendous advances in computer vision. However, despite the stunning performance on some benchmarks, there are several hurdles in bringing these computer vision systems to real world scenarios like robotics. The first hurdle is recognizing small manipulable objects in scenes, where current systems fail because they use bottom-up models that are incapable of top-down reasoning. Another challenge is learning models for rare concepts; including concepts not in our vision datasets and generalizing models to rare instances of known concepts.

In this talk, I will present my efforts to address these limitations and bring computer vision systems closer to working in real world scenarios. I will begin with models that can implicitly learn top-down contextual structure, which is necessary to recognize small and challenging objects. I will then describe a large-scale constrained semi-supervised learning framework, which uses millions of web images to continuously learn models for new concepts and discover relationships between these concepts, without any human annotations. Finally, I will talk about optimization strategies that help algorithms generalize to rare/unusual examples on their own.

View the talk here…

Presenter

- Learn More

Abhinav Shrivastava is a Ph.D. candidate at the Robotics Institute of Carnegie Mellon. He is also working as a research assistant at Google Research since Fall 2016. His research interests are Computer Vision and Machine Learning, and their intersection with fields such as Graphics and Robotics. He received the Microsoft Research Ph.D. Fellowship in 2014. One of his projects, NEIL, was also awarded the top-10 ideas in 2013 by CNN. Several of his works have featured in the popular press, including Forbes, BBC, and the Associated Press.

Details

Date:
February 24, 2017
Time:
11:00 am - 12:00 pm
Event Category: