Human Action Recognition (HAR) aims to understand human behavior and assign a label to
each action. It has a wide range of applications, and therefore has been attracting increasing …

Temporal action segmentation (TAS) in videos aims at densely identifying video frames in
minutes-long videos with multiple action classes. As a long-range video understanding task …

We show, for the first time, that neural networks trained only on synthetic data achieve state-
of-the-art accuracy on the problem of 3D human pose and shape (HPS) estimation from real …

Interactive 3D simulations have enabled break-throughs in robotics and computer vision, but
simulating the broad diversity of environments needed for deep learning requires large …

Assembly101 is a new procedural activity dataset featuring 4321 videos of people
assembling and disassembling 101" take-apart" toy vehicles. Participants work without fixed …

Being able to map the activities of others into one's own point of view is one fundamental
human skill even from a very early age. Taking a step toward understanding this human …

We present a new egocentric procedural error dataset containing videos with various types
of errors as well as normal videos and propose a new framework for procedural error …

The egocentric and exocentric viewpoints of a human activity look dramatically different, yet
invariant representations to link them are essential for many potential applications in …

In order to enable the model to generalize to unseen “action-objects”(compositional action),
previous methods encode multiple pieces of information (ie, the appearance, position, and …

We present a self-supervised approach for learning video representations using temporal
video alignment as a pretext task, while exploiting both frame-level and video-level …