Human pose estimation aims to locate the human body parts and build human body
representation (eg, body skeleton) from input data such as images and videos. It has drawn …

Increased global competition has placed a premium on customer satisfaction, and there is a
greater demand for manufacturers to be flexible with their products and services. This …

We introduce YOLO-pose, a novel heatmap-free approach for joint detection, and 2D multi-
person pose estimation in an image based on the popular YOLO object detection …

In this paper, we are interested in the bottom-up paradigm of estimating human poses from
an image. We study the dense keypoint regression framework that is previously inferior to …

Human pose estimation deeply relies on visual clues and anatomical constraints between
parts to locate keypoints. Most existing CNN-based methods do well in visual …

Current methods of multi-person pose estimation typically treat the localization and
association of body joints separately. In this paper, we propose the first fully end-to-end multi …

Heatmap-based methods dominate in the field of human pose estimation by modelling the
output distribution through likelihood heatmaps. In contrast, regression-based methods are …

While CNN-based models have made remarkable progress on human pose estimation,
what spatial dependencies they capture to localize keypoints remains unclear. In this work …

Deep neural networks based purely on attention have been successful across several
domains, relying on minimal architectural priors from the designer. In Human Action …

High-resolution representations are essential for position-sensitive vision problems, such as
human pose estimation, semantic segmentation, and object detection. Existing state-of-the …