StampNet: unsupervised multi-class object discovery
This addresses the problem of unsupervised object discovery for computer vision researchers, offering a method to handle unknown object locations and categories, though it appears incremental as it builds on autoencoding approaches.
The authors tackled unsupervised multi-class object discovery in images, where object size and location are unknown, by proposing StampNet, an autoencoding neural network that localizes and categorizes objects simultaneously, demonstrating its ability to localize and cluster overlapping shapes including MNIST digits and pedestrians in depth-maps.
Unsupervised object discovery in images involves uncovering recurring patterns that define objects and discriminates them against the background. This is more challenging than image clustering as the size and the location of the objects are not known: this adds additional degrees of freedom and increases the problem complexity. In this work, we propose StampNet, a novel autoencoding neural network that localizes shapes (objects) over a simple background in images and categorizes them simultaneously. StampNet consists of a discrete latent space that is used to categorize objects and to determine the location of the objects. The object categories are formed during the training, resulting in the discovery of a fixed set of objects. We present a set of experiments that demonstrate that StampNet is able to localize and cluster multiple overlapping shapes with varying complexity including the digits from the MNIST dataset. We also present an application of StampNet in the localization of pedestrians in overhead depth-maps.