Scaling up sign spotting through sign language dictionaries

The focus of this work is textitsign spotting - given a video of an isolated sign, our task is to identify textitwhether and textitwhere it has been signed in a continuous, co-articulated sign language video. To achieve this sign spotting task, we train a model using multiple types of available supervision by: (1) textitwatching existing footage which is sparsely labelled using mouthing cues; (2) textitreading associated subtitles (readily available translations of the signed content) which provide additional textitweak-supervision; (3) textitlooking up words (for which no co-articulated labelled examples are available) in visual sign language dictionaries to enable novel sign spotting. These three tasks are integrated into a unified learning framework using the principles of Noise Contrastive Estimation and Multiple Instance Learning. We validate the effectiveness of our approach on low-shot sign spotting benchmarks. In addition, we contribute a machine-readable British Sign Language (BSL) dictionary dataset of isolated signs, BSLDict, to facilitate study of this task. The dataset, models and code are available at our project page.