Evaluating the Immediate Applicability of Pose Estimation for Sign Language Recognition
Authors:
- 1 June 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3429-3435
Abstract
Sign languages are visual languages produced by the movement of the hands, face, and body. In this paper, we evaluate representations based on skeleton poses, as these are explainable, person-independent, privacy-preserving, low-dimensional representations. Basically, skeletal representations generalize over an individual’s appearance and background, allowing us to focus on the recognition of motion. But how much information is lost by the skeletal representation? We perform two independent studies using two state-of-the-art pose estimation systems. We analyze the applicability of the pose estimation systems to sign language recognition by evaluating the failure cases of the recognition models. Importantly, this allows us to characterize the current limitations of skeletal pose estimation approaches in sign language recognition.Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Effective Approaches to Attention-based Neural Machine Translation2015
- Automatic and Efficient Human Pose Estimation for Sign Language VideosInternational Journal of Computer Vision, 2013
- Head Pose Estimation for Sign Language VideoLecture Notes in Computer Science, 2013
- Articulated people detection and pose estimation: Reshaping the future2012
- Automatic and Efficient Long Term Arm and Hand Tracking for Continuous Sign Language TV Broadcasts2012
- American sign language recognition with the kinect2011
- Real-time human pose recognition in parts from single depth images2011
- Sign Language Recognition2011
- Connectionist temporal classification2006
- Individual Comparisons by Ranking Methods1992