Adding SignBERT+

[cleong110]

Adding SignBERT+ summary

Paper: https://ieeexplore.ieee.org/document/10109128

Detailed description of summary process: https://github.com/cleong110/sign-language-processing.github.io/issues/14

Summary of PR:

• Adding the paper and summary
• Adding some key references

Questions:

• should this be in the pose-to-text section? That is one of the tasks they evaluate on but it's a pretraining framework for general SLU, and they just validate on it.
• Would it be good to have a general "pretraining" section instead?
• Added a citation for RWTH-PHOENIX-WEATHER (the 2015 continuous SLR set, koller2015ContinuousSLR, not the translation set!), because it seemed we didn't have one. Do we need an explanation somewhere of the distinction between all the different RWTH sets?
• (lower priority) What do we do about the HANDS17 dataset or other similar datasets that aren't specifically for SLU but are useful?
• What do we do about SignBERT (https://openaccess.thecvf.com/content/ICCV2021/html/Hu_SignBERT_Pre-Training_of_Hand-Model-Aware_Representation_for_Sign_Language_Recognition_ICCV_2021_paper.html)? They describe SignBERT+ as an extension of SignBERT, should we mention it?