Abstract | ||
---|---|---|
Motion recognition systems have been widely developed in the field of human computer interaction. Methods, such as pointing, dynamic gesture and static gesture or hand held devices have been proposed for motion recognition. The motion recognition systems have been gradually adapted to home appliances in our daily. In this paper, we focus on TV interaction, since the device is a recent representative multimedia device applying the motion technique. Most motion recognition systems utilize 3D data, such as horizontal, vertical and depth information by stereo camera or ToF (Time of Flight) camera. However, this paper proposes the different techniques for human-TV interaction. We propose an optical flow based motion recognition system that provides direction and speed, in addition to the position of the moving target in real time. These factors are useful in recognizing human motion more effectively and more dynamically. Therefore, we design the natural interaction for human-TV using these motion data. The calculation process of optical flow is outside the scope of this paper. This real time optical flow calculation is implemented using the FPGA chip supporting parallel processing by a hardware team in our laboratory. We propose a method for human motion recognition based on real time optical flow system. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1007/978-3-642-31137-6_24 | ICCSA (3) |
Keywords | Field | DocType |
human motion,motion recognition,motion data,tv interaction,human hand motion,motion technique,human motion recognition,motion recognition system,optical flow system,human computer interaction,optical flow,tv remote control,real time | Stereo camera,Computer vision,Remote control,Motion recognition,Computer science,Gesture,Real-time operating system,Artificial intelligence,Motion estimation,Optical flow,Match moving | Conference |
Volume | ISSN | Citations |
7335 | 0302-9743 | 3 |
PageRank | References | Authors |
0.40 | 10 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Soonmook Jeong | 1 | 24 | 3.42 |
Taehoun Song | 2 | 23 | 3.04 |
Keyho Kwon | 3 | 11 | 3.05 |
Jae Wook Jeon | 4 | 565 | 77.35 |