Spatio-Temporal Pyramid Graph Convolutions for Human Action Recognition and Postural Assessment

Conference paper

Recognition of human actions and associated interactions with objects and the environment is an important problem in computer vision due to its potential applications in a variety of domains. Recently, graph convolutional networks that extract features from the skeleton have demonstrated promising performance. In this paper, we propose a novel Spatio-Temporal Pyramid Graph Convolutional Network (ST-PGN) for online action recognition for ergonomics risk assessment that enables the use of features from all levels of the skeleton feature hierarchy. The proposed algorithm outperforms state-of-art action recognition algorithms tested on two public benchmark datasets typically used for postural assessment (TUM and UW-IOM). We also introduce a pipeline to enhance postural assessment methods with online action recognition techniques. Finally, the proposed algorithm is integrated with a traditional ergonomics risk index (REBA) to demonstrate the potential value for assessment of musculoskeletal disorders in occupational safety.

Activities

Selected Works

Awards

Selected Accomplishments

Here is where I collect some of my accomplishments.

Read Me

Media

Published Articles on my Work

Here is where I collect articles that are published about my work.

Read Me

SmartErgo App

Mobile App Design

SmartErgo is a mobile app that identifies ergonomically unsafe actions.

Read Me

Publication

Conference paper

Spatio-Temporal Pyramid Graph Convolutions for Human Action Recognition and Postural Assessment

Read Me

Curiosity Day

Outreach Program

I volunteered at the Pacific Science Center curiosity day, and made a video to teach kids how optical tweezers work.

Read Me