Why interesting? Language is abstractive and videos are dense, while sign language bridges low-level'' vision and
high-level'' language. Eveyone could somehow understand sign language.
It is worth noting that different culture also raise difference. For instance, Shaking heads mean ``Yes'' in India.
https://dxli94.github.io/WLASL/
American English
2000 Sign Classes by 119 Signers. 21K Samples
http://home.ustc.edu.cn/~alexhu/Sources/index.html
Chinese
1067 Sign Classes by 10 Signers. 32K Samples
https://www.robots.ox.ac.uk/~vgg/data/bobsl/
British English
2281 Sign Classes by 39 Signers. 452K Samples
StepNet: Spatial-temporal Part-aware Network for Sign Language Recognition. arXiv 2022
Signbert: Pre-training of hand-model-aware representation for sign language recognition. ICCV 2021
Global-local enhancement network for nmf-aware sign language recognition. ACM TOMM 2021
A deep neural framework for continuous sign language recognition by iterative training. TMM 2019
Spatial temporal graph convolutional networks for skeleton-based action recognition. AAAI 2018
Jointly Harnessing Prior Structures and Temporal Consistency for Sign Language Video Generation. arXiv 2022