A dataset representing the movements that a person must perform to produce a sign

It includes both static and dynamic signs, represented as 3D coordinate data of the face, torso and hands.The dataset is organized into a single JSON file with six hierarchical levels, each corresponding to:·        signer number: The identifier assigned to each participant.·        sign topic: The semantic category of the sign, which in this case includes colors, numbers, and courtesy expressions.·        sign name: The word represented by the sign.·        video number: The identifier for the original video recording of a given sign.·        repetition number: The identifier for each augmented instance derived from a single video.·        frame number: The sequential index of each frame, which contains the spatial coordinates.Additionally, all raw videos are included in the 'DataSet' folder, which contains three ZIP files, each corresponding to a directory of sign videos for the categories: colors, courtesy, and numbers.