Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Improving Continuous Sign Language Recognition with Consistency Constraints and Signer Removal

About

Most deep-learning-based continuous sign language recognition (CSLR) models share a similar backbone consisting of a visual module, a sequential module, and an alignment module. However, due to limited training samples, a connectionist temporal classification loss may not train such CSLR backbones sufficiently. In this work, we propose three auxiliary tasks to enhance the CSLR backbones. The first task enhances the visual module, which is sensitive to the insufficient training problem, from the perspective of consistency. Specifically, since the information of sign languages is mainly included in signers' facial expressions and hand movements, a keypoint-guided spatial attention module is developed to enforce the visual module to focus on informative regions, i.e., spatial attention consistency. Second, noticing that both the output features of the visual and sequential modules represent the same sentence, to better exploit the backbone's power, a sentence embedding consistency constraint is imposed between the visual and sequential modules to enhance the representation power of both features. We name the CSLR model trained with the above auxiliary tasks as consistency-enhanced CSLR, which performs well on signer-dependent datasets in which all signers appear during both training and testing. To make it more robust for the signer-independent setting, a signer removal module based on feature disentanglement is further proposed to remove signer information from the backbone. Extensive ablation studies are conducted to validate the effectiveness of these auxiliary tasks. More remarkably, with a transformer-based backbone, our model achieves state-of-the-art or competitive performance on five benchmarks, PHOENIX-2014, PHOENIX-2014-T, PHOENIX-2014-SI, CSL, and CSL-Daily. Code and Models are available at https://github.com/2000ZRL/LCSA_C2SLR_SRM.

Ronglai Zuo, Brian Mak• 2022

Related benchmarks

TaskDatasetResultRank
Continuous Sign Language RecognitionPHOENIX 2014 (dev)
Word Error Rate20.5
188
Continuous Sign Language RecognitionPHOENIX-2014 (test)
WER20.4
185
Continuous Sign Language RecognitionCSL-Daily (dev)
Word Error Rate (WER)31.9
98
Continuous Sign Language RecognitionCSL-Daily (test)
WER31
91
Continuous Sign Language RecognitionPHOENIX14-T (dev)
WER20.2
75
Continuous Sign Language RecognitionCSL (test)
WER0.68
23
Continuous Sign Language RecognitionPHOENIX-2014-SI (test)
WER32.7
5
Showing 7 of 7 rows

Other info

Code

Follow for update