Abstract

 
 
 This work addresses the problem of recognizing the American Sign Language (ASL) hand alphabet relying only on depth information acquired from an RGB-D sensor. To accomplish this goal, a novel Iterative Closest Point (ICP) based recognition methodology is proposed where it comprehensively analyzes the inputs and outputs of the alignment as efficiency and accuracy determinants. Next, a novel classification technique, denoted Approximated KB-fit, is proposed to efficiently handle the space complexity of the database template matching. The overall accuracy of the recognition reached a performance of 99.04% in a cross-validation workbench with 520 distinct input depth images. The achieved frame rate was 7.41 FPS performed on a 2.4 GHz single processor based machine.
 
 
Highlights
The recent introduction of low-cost sensor devices, empowered with real-time RGB-D image acquisition mechanisms, favored the appearance of many innovative computer vision works
The results presented show that the Iterative Closest Point (ICP) algorithm can be used to produce accurate matches even with a very similar set of gestures poses
As ICP processing is always conditioned to the pairwise data alignments, the general template matching paradigm is still a bottleneck to its application in real-time contexts (≈ 15 FPS)
Summary
The recent introduction of low-cost sensor devices, empowered with real-time RGB-D image acquisition mechanisms, favored the appearance of many innovative computer vision works. One can find accurate and efficient solutions in the literature, they often require additional hardware or accessories which may be expensive or demand a complex setup. Iterative Closest Point (ICP) [10] is the dominant fine registration algorithm in the literature and it aims at the retrieval of an accurate solution to the Euclidean rigid motion between two 3D point surfaces. The rigid motion (transformation) is usually recovered by means of the scale, rotation and translation components that bring the two point sets to a same spatial orientation. This way, the ICP algorithm works by iteratively minimizing the cost function of the distances computed between selected corresponding points in the two surfaces (Figure 3). Along one iterative step, the method needs to search for another set of points on the second model, mi ∈ M , which minimizes the distance cost function between T (pi) and mi (Equation 1)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.