Most of the sign language recognition (SLR) systems rely on supervision for training and available annotated sign language resources are scarce due to the difficulties of manual labeling. Unsupervised discovery of lexical units would facilitate the annotation process and thus lead to better SLR systems. Inspired by the unsupervised spoken term discovery in speech processing field, we investigate whether a similar approach can be applied in sign language to discover repeating lexical units. We adapt an algorithm that is designed for spoken term discovery by using hand shape and pose features instead of speech features. The experiments are run on a large scale continuous sign corpus and the performance is evaluated using gloss level annotations. This work introduces a new task for sign language processing that has not been addressed before.

Sign language research most often relies on exhaustively annotated and segmented data, which is scarce even for the most studied sign languages. However, parallel corpora consisting of sign language interpreting are rarely explored. By utilizing such data for the task of keyword search, this work aims to enable information retrieval from sign language with the queries from the translated written language. With the written language translations as labels, we train a weakly supervised keyword search model for sign language and further improve the retrieval performance with two context modeling strategies. In our experiments, we compare the gloss retrieval and cross language retrieval performance on RWTH-PHOENIX-Weather 2014T dataset.