Deep Set Prediction Networks
This addresses a fundamental issue in machine learning for tasks involving set prediction, such as object detection and attribute recognition, though it appears incremental as it builds on prior set-based approaches.
The paper tackled the problem of predicting unordered sets from feature vectors, which existing methods mishandle, and proposed a general model that properly respects set structure, demonstrating its ability to auto-encode point sets and predict bounding boxes and attributes in images.
Current approaches for predicting sets from feature vectors ignore the unordered nature of sets and suffer from discontinuity issues as a result. We propose a general model for predicting sets that properly respects the structure of sets and avoids this problem. With a single feature vector as input, we show that our model is able to auto-encode point sets, predict the set of bounding boxes of objects in an image, and predict the set of attributes of these objects.