CASL-W60

About DatasetThis is a dataset of Central Africa Sign Language (CASL), a visual-gestural language within the Central African region. This dataset is among the very first of its kind to cater for a section of CASL in terms of public availability for computational modeling and linguistic analysis.The data was collected from 19 volunteers who performed signs following a reference video of an expert in CASL. The volunteers contrasted in diverse aspects such as age, gender, and demographic variations.The dataset consists of 60 commonly used words and signs, which include various types of expressions, especially nouns and verbs. Each participant signed each word at least 10 times, resulting in 5889 video samples. The recordings were done using the same camera, which was computer webcam, in different environments, both indoor and outdoor, on a non-uniform background with variations in lighting.

AcknowledgementsAll the CASL-W60 data is intended for academic and computational use only. No commercial usage is allowed. Licensed under the Computational Use of Data Agreement (C-UDA). The full agreement can be accessed here (https://news.microsoft.com/wp-content/uploads/prod/sites/560/2019/12/C-UDA-1.0.pdf)

A Data Usage Agreement (DUA) document is also uploaded for the researchers that needs to be agreed upon to use the CASL-W60 dataset.

Data accessibilityThe dataset is available for peer review only under the C-UDA license. By downloading this dataset, users acknowledge and agree to adhere to the license terms. Any further query related to the dataset should be communicated through [e-mail].