Visual Alignment Constraint for Continuous Sign Language Recognition

Vision-based Continuous Sign Language Recognition (CSLR) aims to recognize unsegmented signs from image streams. Overfitting is one of the most critical problems in CSLR training, and previous works show that the iterative training scheme can partially solve this problem while also costing more trai...

Full description

Saved in:
Bibliographic Details
Main Authors: Min, Yuecong, Hao, Aiming, Chai, Xiujuan, Chen, Xilin
Format: Conference Proceeding
Language:English
Subjects:
Action and behavior recognition
Assistive technologies
Costing
Gesture recognition
Gestures and body pose
Image recognition
Measurement
Training
Visualization
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Vision-based Continuous Sign Language Recognition (CSLR) aims to recognize unsegmented signs from image streams. Overfitting is one of the most critical problems in CSLR training, and previous works show that the iterative training scheme can partially solve this problem while also costing more training time. In this study, we revisit the iterative training scheme in recent CSLR works and realize that sufficient training of the feature extractor is critical to solving the overfitting problem. Therefore, we propose a Visual Alignment Constraint (VAC) to enhance the feature extractor with alignment supervision. Specifically, the proposed VAC comprises two auxiliary losses: one focuses on visual features only, and the other enforces prediction alignment between the feature extractor and the alignment module. Moreover, we propose two metrics to reflect overfitting by measuring the prediction inconsistency between the feature extractor and the alignment module. Experimental results on two challenging CSLR datasets show that the proposed VAC makes CSLR networks end-to-end trainable and achieves competitive performance.
ISSN:2380-7504
DOI:10.1109/ICCV48922.2021.01134