Description

Data was captured using a setup that consisted of: - Two Fifth Dimension Technologies (5DT) gloves, one right and one left - Two Ascension Flock-of-Birds magnetic position trackers, one attached to each hand - A four-port serial card to cope with four data sources - A PC (128MB RAM, Intel Pentium II 266MHz) was used In terms of the quality of the data, the Flock system was far superior to the Nintendo system also available from the same donor. Firstly, this was a two-hand system. Secondly, each position tracker provided 6 degrees of freedom - i.e. roll, pitch and yaw as well as x, y and z. The gloves also provided a full five fingers of data. But the big improvements were in resolution - both accuracy and temporal. Position and orientation were defined to 14-bit accuracy, giving position information with a typical positional error less than one centimetre and angle error less than one half of a degree. Finger bend was measured with 8 bits per finger, of which probably 6 bits were usable once the glove was calibrated. The refresh rate of the complete system was close to 100 frames per second; and all signals had significantly less noise than the Nintendo data. Samples from a single signer (a native Auslan signer) were collected over a period of nine weeks. In total, 27 samples per sign, and a total of 2565 signs were collected. The average length of each sign was approximately 57 frames. The data was collected from a volunteer native Auslan signer The data presented is the raw data with no filtering. The file consists of 9 subdirectories tctodd1-9. Each directory consists of 3 samples of each sign, captured on a different day. In total there are 95 different signs, with 27 samples per sign. Signs were provided by a native signer volunteer. Each file consists of a sequence of lines. Each line consists of 22 whitespace-separated numbers representing the 22 channels of information. The list of channels can be found in the domain description file. It also lists the classes. More information can be found here: [Web Link].

Related Papers

  • Mohammed Waleed Kadous and Claude Sammut. The University of New South Wales School of Computer Science and Engineering Temporal Classification: Extending the Classification Paradigm to Multivariate Time Series. [link]
  • [link]
  • [link]

Related datasets