Learning from Incomplete and Heterogeneous Data
Abstract
Over the last decade, methods based on Deep Convolutional Neural Networks (DCNNs) have shown impressive performance improvements for object detection and recognition problems. This has been made possible due to the availability of large annotated datasets, a better understanding of the non-linear mapping between input images and class labels as well as the affordability of GPUs. However, a vast majority of DCNN-based recognition methods are designed for a closed world, where the primary assumption is that all categories are known a priori. In many real-world applications, this assumption does not necessarily hold. Generally, incomplete knowledge of the world is present at training time, and unknown classes can be submitted to an algorithm during testing. The goal of a visual recognition system is then to reject samples from unknown classes and classify samples from known classes. This project aims to overcome fundamental limitations of the current DCNNs for visual recognition in rejecting unknown class samples. Specific goals of this research are to (1) develop DCNN-based methods for anomaly detection; (2) develop novel open-set recognition networks; (3) develop domain adaptive anomaly detection and open-set recognition methods that can deal with changing distribution of samples; and (4) develop methods that can defend anomaly detection and open-set recognition networks against adversarial attacks. Applications of this research include image classification and automatic target recognition.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 25, 2021
- Source ID
- W911NF2110135
Entities
People
- Vishal Patel
Organizations
- Army Contracting Command
- Johns Hopkins University
- United States Army