Data Set Information:

该数据集由从7个手势视频中提取的特征组成，旨在研究手势相位分割。

每个视频由两个文件表示：一个原始文件，其中包含用户在每个帧中的手、手腕、头部和脊椎的位置；以及一个经过处理的文件，其中包含手和手腕的速度和加速度。有关数据集的详细信息，请参见数据集说明。

Attribute Information:

Raw files: 18 numeric attributes (double), a timestamp and a class attribute (nominal).
Processed files: 32 numeric attributes (double) and a class attribute (nominal).
A feature vector with up to 50 numeric attributes can be generated with the two files mentioned above.

Relevant Papers:

1.  Madeo, R. C. B. ; Lima, C. A. M. ; PERES, S. M. . Gesture Unit Segmentation using Support Vector Machines: Segmenting
      Gestures from Rest Positions. In: Symposium on Applied Computing (SAC), 2013, Coimbra. Proceedings of the 28th Annual
      ACM Symposium on Applied Computing (SAC), 2013. p. 46-52.
      * In this paper, the videos A1 and A2 were studied.

  2.  Wagner, P. K. ; PERES, S. M. ; Madeo, R. C. B. ; Lima, C. A. M. ; Freitas, F. A. . Gesture Unit Segmentation Using
      Spatial-Temporal Information and Machine Learning. In: 27th Florida Artificial Intelligence Research Society Conference
      (FLAIRS), 2014, Pensacola Beach. Proceedings of the 27th Florida Artificial Intelligence Research Society Conference
      (FLAIRS). Palo Alto : The AAAI Press, 2014. p. 101-106.
      * In this paper, the videos A1, A2, A3, B1, B3, C1 and C3 were studied.

  3.  Madeo, R. C. B.. Support Vector Machines and Gesture Analysis: incorporating temporal aspects  (in Portuguese). Master
      Thesis - Universidade de Sao Paulo, Sao Paulo Researcher Foundation. 2013.
      * In this document, the videos named B1 and B3 in the document correspond to videos C1 and C3 in this dataset. only
      five videos were explored in this document: A1, A2, A3, C1 and C3.

  4.  Wagner, P. K. ; Madeo, R. C. B. ; PERES, S. M. ; Lima, C. A. M. . Segmenta?§ao de Unidades Gestuais com Multilayer
      Perceptrons (in Portuguese). In: Encontro Nacional de Inteligencia Artificial e Computacional (ENIAC), 2013, Fortaleza.
      Anais do X Encontro Nacional de Inteligencia Artificial e Computacional (ENIAC), 2013.
      * In this paper, the videos A1, A2 and A3 were studied.

Citation Request:

Please refer to the Machine Learning Repository's citation policy.
Additionally, the authors require a citation to one or more publications from those cited as relevant papers.

Creators:
    Renata Cristina Barros Madeo (Madeo, R. C. B.)
    Priscilla Koch Wagner (Wagner, P. K.)
    Sarajane Marques Peres (Peres, S. M.)
    {renata.si, priscilla.wagner, sarajane} at usp.br
    http://each.uspnet.usp.br/sarajane/

Donor:
    University of Sao Paulo - Brazil