Event Location: NSH 3305
Bio: Yuxiong Wang is a PhD student in the Robotics Institute at Carnegie Mellon University, where he is supervised by Prof. Martial Hebert. His research focuses on computer vision and machine learning and he is particularly interested in object recognition problems.

Abstract: In this work, we explore an approach to generating detectors for a new detection task that is radically different from the conventional approach of learning a detector from a large corpus of annotated positive and negative data samples. Instead, we assume that we have evaluated off-line a large library of detectors against a large set of detection tasks. Given a new input task, we evaluate a subset of the models on samples from the new task and use the huge matrix of models-tasks ratings to predict the performance of all the models in the library on the new task, enabling us to select a good set of detectors for the new task.

This approach has two key advantages of great interest in practice: 1) a far smaller set of annotated samples is needed compared to the size of the training sets required for training from scratch; and 2) recommending models becomes a very fast operation compared to the notoriously expensive training procedures of modern detectors. (1) will dramatically reduce the need for manually annotating vast datasets for training detectors; (2) will enable rapid generation of new detectors.