DeepSetNet: Predicting Sets with Deep Neural Networks
This work addresses the challenge of predicting sets with variable sizes and unordered elements in computer vision, offering a novel approach that improves performance in tasks like image classification and object detection, though it is incremental as it builds on existing deep learning methods.
The paper tackles the problem of set prediction in deep learning, which is crucial for tasks like image tagging and object detection where outputs are sets with variable sizes and unordered elements. It introduces a likelihood for set distributions and a cardinality loss, achieving superior performance over existing methods in multi-class image classification, object counting, and pedestrian detection on standard datasets.
This paper addresses the task of set prediction using deep learning. This is important because the output of many computer vision tasks, including image tagging and object detection, are naturally expressed as sets of entities rather than vectors. As opposed to a vector, the size of a set is not fixed in advance, and it is invariant to the ordering of entities within it. We define a likelihood for a set distribution and learn its parameters using a deep neural network. We also derive a loss for predicting a discrete distribution corresponding to set cardinality. Set prediction is demonstrated on the problem of multi-class image classification. Moreover, we show that the proposed cardinality loss can also trivially be applied to the tasks of object counting and pedestrian detection. Our approach outperforms existing methods in all three cases on standard datasets.