1

We propose a novel framework to jointly learn the affinity graph and low-rank tensor decomposition for multi-view clustering.

Monocular object pose estimation is an important yet challenging computer vision problem. Depth features can provide useful information for pose estimation. However, existing methods rely on real depth images to extract depth features, leading to its …

Inspired from the dynamical systems, this study aims to unravel and improve ResNets.

We propose a novel training algorithm to improve the generalization performance of GCNs on graphs with few labeled nodes.

Our study aims to explicitly predict overlap relations and resolve overlaps in a differentiable way for the panoptic output.

Fisher’s criterion is one of the most popular discriminant criteria for feature extraction. It is defined as the generalized Rayleigh quotient of the between-class scatter distance to the within-class scatter distance. Consequently, Fisher’s …

Training accurate deep neural networks (DNNs) in the presence of noisy labels is an important and challenging task. Though a number of approaches have been proposed for learning with noisy labels, many open issues remain. In this paper, we show that …

We formulate the attention mechanism into an expectation-maximization manner and iteratively estimate a much more compact set of bases upon which the attention maps are computed.

By rewriting the activation function as an equivalent proximal operator, we approximate a feed-forward neural network by adding the proximal operators to the objective function as penalties, hence we call the lifted proximal operator machine (LPOM).

To achieve simultaneously feature learning and subspace clustering, we propose an end-to-end trainable framework called the Self-Supervised Convolutional Subspace Clustering Network (S2ConvSCN) that combines a ConvNet module (for feature learning), a self-expression module (for subspace clustering) and a spectral clustering module (for self-supervision) into a joint optimization framework.