Exjobbsförslag från företag

Detta är ett uppsatsförslag hämtat från Nationella Exjobb-poolen. Klicka här för att komma tillbaka till samtliga exjobbsförslag.

Förslaget inkom 2011-04-21

Robust Image-based Recognition of Sign Language Signs

OBS! ANSÖKNINGSTIDEN FÖR DETTA EXJOBB HAR LÖPT UT.
Automatic recognition of sign language is a research area pertaining to several different areas in computer science, such as computer vision and language technology. Sign language technology research has attracted a lot of attention recently, with a potential to dramatically improve accessibility in society much in the same way as speech technology has in recent years [1].
The proposed project aims to investigate a novel approach for recognition of individual sign language signs from video and/or 3D sensor input (e.g. Microsoft Kinect). The method will build on a fast and robust hand pose estimation method developed at KTH CVAP [2]. The basic idea in the method is to represent the recognition problem as one-of-search. A very large number of images of hands are generated using the graphics software Poser™. A set of parameters (the joint angles of the hand – or in the sign language case, the sign type) is attached to each image. Given a new image, hand images with similar appearance are retrieved using an approximate nearest neighbor lookup method. The pose (or in the sign language case, the sign type) of the new hand is estimated as the weighted mean of the poses (sign types) of the retrieved nearest neighbors. The project will utilize existing sign language video resources for training and testing such as Svenskt Teckenspråkslexikon from SU and http://www.spreadthesign.com.
Possibility of funded extension of the thesis project.


[1] Beskow, J., & Granström, B. (2010). Teckenspråkteknologi - sammanfattning av förstudie. Technical Report, KTH Centrum för Talteknologi.
[2] Javier Romero, Hedvig Kjellström and Danica Kragic. Hands in action: Real-time 3D reconstruction of hands in interaction with objects. In IEEE International Conference on Robotics and Automation, 2010.

keywords: [email protected], [email protected]

  GÅ TILL XJOBB.NU FÖR FULLSTÄNDIG INFO OM DETTA EXJOBB




Informationen om uppsatsförslag är hämtad från Nationella Exjobb-poolen.