HandReader: Advanced Techniques for Efficient Fingerspelling Recognition

Open paper
Benchmark Model Rank Results
sign-language-recognition-on-chicagofswildHandReader_RGB_KP#1CER (%): 27.1
sign-language-recognition-on-chicagofswildHandReader_KP#2CER (%): 28
sign-language-recognition-on-chicagofswildHandReader_RGB#3CER (%): 30.7
sign-language-recognition-on-chicagofswild-plusHandReader_RGB+KP#1CER (%): 24.4
sign-language-recognition-on-chicagofswild-plusHandReader_KP#2CER (%): 26.2
sign-language-recognition-on-chicagofswild-plusHandReader_RGB#3CER (%): 27.6
sign-language-recognition-on-znakiHandReader_RGB_KP#1CER (%): 5.06
sign-language-recognition-on-znakiHandReader_KP#2CER (%): 7.35
sign-language-recognition-on-znakiHandReader_RGB#3CER (%): 7.61

Rank counts only results with a code link.