Object category learning and retrieval with weak supervision
This work addresses the challenge of object categorization in computer vision with weak supervision, offering an incremental improvement by integrating clustering and feature learning in a differentiable framework.
The paper tackles the problem of learning and retrieving object categories from images with minimal supervision, proposing an unsupervised deep clustering method that achieves promising results on CIFAR10 and Cityscapes datasets for discovering semantic classes like cars, people, and bicycles.
We consider the problem of retrieving objects from image data and learning to classify them into meaningful semantic categories with minimal supervision. To that end, we propose a fully differentiable unsupervised deep clustering approach to learn semantic classes in an end-to-end fashion without individual class labeling using only unlabeled object proposals. The key contributions of our work are 1) a kmeans clustering objective where the clusters are learned as parameters of the network and are represented as memory units, and 2) simultaneously building a feature representation, or embedding, while learning to cluster it. This approach shows promising results on two popular computer vision datasets: on CIFAR10 for clustering objects, and on the more complex and challenging Cityscapes dataset for semantically discovering classes which visually correspond to cars, people, and bicycles. Currently, the only supervision provided is segmentation objectness masks, but this method can be extended to use an unsupervised objectness-based object generation mechanism which will make the approach completely unsupervised.