Weakly-supervised Discovery of Visual Pattern Configurations
This addresses the need for object detection methods that can work with minimal supervision, which is incremental but improves performance in a specific domain.
The paper tackles the problem of weakly-supervised object detection by automatically identifying discriminative visual pattern configurations, leading to state-of-the-art results on the PASCAL VOC dataset.
The increasing prominence of weakly labeled data nurtures a growing demand for object detection methods that can cope with minimal supervision. We propose an approach that automatically identifies discriminative configurations of visual patterns that are characteristic of a given object class. We formulate the problem as a constrained submodular optimization problem and demonstrate the benefits of the discovered configurations in remedying mislocalizations and finding informative positive and negative training examples. Together, these lead to state-of-the-art weakly-supervised detection results on the challenging PASCAL VOC dataset.