| | |
Summary: Recognizing Hand Gesture Using Motion Trajectories
Ming-Hsuan Yang and Narendra Ahuja
Beckman Institute and Department of Electrical and Computer Engineering
University of Illinois at Urbana-Champaign, Urbana, IL 61801
fmhyang,ahujag@vision.ai.uiuc.edu
http://vision.ai.uiuc.edu
Abstract
We present an algorithm for extracting and clas-
sifying two-dimensional motion in an image sequence
based on motion trajectories. First, a multiscale seg-
mentation is performed to generate homogeneous re-
gions in each frame. Regions between consecutive
frames are then matched to obtain 2-view correspon-
dences. AĆne transformations are computed from
each pair of corresponding regions to de ne pixel
matches. Pixels matches over consecutive images pairs
are concatenated to obtain pixel-level motion trajecto-
ries across the image sequence. Motion patterns are
learned from the extracted trajectories using a time-
delay neural network. We apply the proposed method
|
