3D human pose and shape estimation from monocular images has been an active research area in computer vision. Existing deep learning methods for this task rely on high-resolution input, which however,

3D human shape and pose estimation from monocular images has been an active area of research in computer vision, having a substantial impact on the development of new applications, from activity recog

Joint rolling shutter correction and deblurring (RSCD) techniques are critical for the prevalent CMOS cameras. However, current approaches are still based on conventional energy optimization and are d

We consider the problem of optimization of deep learning models with smooth activation functions. While there exist influential results on the problem from the ``near initialization'' perspective, we

Metamagnetic transitions in the novel spin-triplet superconductor UTe$_2$ were investigated through the newly developed simultaneous measurements of magnetization and sample temperature for the field

We identify and document a new principle of economic behavior: the principle of the Malevolent Hiding Hand. In a famous discussion, Albert Hirschman celebrated the Hiding Hand, which he saw as a benev

The author suggests a low cost special AB-Net from artificial fiber, which may protect cities and important objects from rockets, artillery and mortar shells, projectiles, bullets, and strategic weapo

The ultimate goal of video prediction is not forecasting future pixel-values given some previous frames. Rather, the end goal of video prediction is to discover valuable internal representations from

Human matting refers to extracting human parts from natural images with high quality, including human detail information such as hair, glasses, hat, etc. This technology plays an essential role in ima

In recent years, many works in the video action recognition literature have shown that two stream models (combining spatial and temporal input streams) are necessary for achieving state of the art per

Purpose: To introduce a dual-domain reconstruction network with V-Net and K-Net for accurate MR image reconstruction from undersampled k-space data. Methods: Most state-of-the-art reconstruction metho

We present the Grasp Proposal Network (GP-net), a Convolutional Neural Network model which can generate 6-DOF grasps for mobile manipulators. To train GP-net, we synthetically generate a dataset conta

We introduce a fast and simple method of computing cumulants of net-proton or net-charge fluctuations in event-by-event hydrodynamic simulations of heavy-ion collisions. One evaluates the mean numbers

The anti-interference capability of wireless links is a physical layer problem for edge computing. Although convolutional codes have inherent error correction potential due to the redundancy introduce

Site-Net is a transformer architecture that models the periodic crystal structures of inorganic materials as a labelled point set of atoms and relies entirely on global self-attention and geometric in