Sorry, you need to enable JavaScript to visit this website.

The manuscript describes fast and scalable architectures and associated algorithms for computing convolutions and cross-correlations. The basic idea is to map 2D convolutions and cross-correlations to a collection of 1D convolutions and cross-correlations in the transform domain. This is accomplished through the use of the Discrete Periodic Radon Transform (DPRT) for general kernels and the use of SVD-LU decompositions for low-rank kernels.

This work presents an automatic method for optical flow inpainting. Given a video, each frame domain is endowed with a Riemannian metric based on the video pixel values. The missing optical flow is recovered by solving the Absolutely Minimizing Lipschitz Extension (AMLE) partial differential equation on the Riemannian manifold. An efficient numerical algorithm is proposed using eikonal operators for nonlinear elliptic partial differential equations on a finite graph.

In this paper, we propose a novel approach for the domain of makeup-robust face recognition. Most face recognition schemes usually fail to generalize well on these data where there is a large difference between the training and testing sets, e.g., makeup changes. Our method focuses on the problem of determining whether face images before and after makeup refer to the same identity. The work on this fundamental research topic benefits various real-world applications, for example automated passport control, security in general, and surveillance.

Visual attention allows the human visual system to effectively deal with the huge flow of visual information acquired by the retina. Since the years 2000, the human visual system began to be modelled in computer vision to predict abnormal, rare and surprising data. Attention is a product of the continuous interaction between bottom-up (mainly feature-based) and top-down (mainly learning-based) information. Deep-learning (DNN) is now well established in visual attention modelling with very effective models.

In this work, we address human action recognition problem under viewpoint variation. The proposed model is formulated by wisely combining convolution neural network (CNN) model with principle component analysis (PCA). In this context, we pass real depth videos through a CNN model in a frame-wise manner. The view invariant features are extracted by employing convolution layers as mid-outputs and considered as 3D nonnegative tensors.

In this work, we address human action recognition problem under viewpoint variation. The proposed model is formulated by wisely combining convolution neural network (CNN) model with principle component analysis (PCA). In this context, we pass real depth videos through a CNN model in a frame-wise manner. The view invariant features are extracted by employing convolution layers as mid-outputs and considered as 3D nonnegative tensors.

This paper introduces architecture with high throughput, low on-chip memory, and efficient data access for Improved Dense Trajectories (iDT) as video representations for real-time action recognition. The iDT feature can capture long-term motion cues better than any existing deep feature, which makes it crucial in state-of-the-art action recognition systems.