Techniques for determining sign language gesture partially shown in image(s)

A method, comprising:

• providing, at a first device, at least one image showing a first gesture into a gesture classifier to receive, as output from the gesture classifier, a first text word corresponding to the first gesture;
• providing, at the first device, at least one image partially but not fully showing a second gesture into the gesture classifier to receive, as output from the gesture classifier, a second text word corresponding to the second gesture, the second gesture being different from the first gesture, the second text word being different from the first text word;
• and
• providing, to an apparatus different from the first device, a text string indicating the first text word and the second text word.

At least one computer readable storage medium (CRSM) that is not a transitory signal, the computer readable storage medium comprising instructions executable by at least one processor to:

• receive one or more images at a first device, the one or more images indicating a first gesture being made by a person using a sign language, at least part of the first gesture extending out of at least one respective image frame of the one or more images;
• provide the one or more images to a gesture classifier;
• receive, from the gesture classifier, plural candidate first text words for the first gesture;
• use context determined from at least a second text word that has been correlated to a second gesture different from the first gesture to select one of the candidate first text words for the first gesture;
• combine the second text word with the selected first text word to establish a text string;
• and
• provide the text string to an apparatus.