Feature Fusion using Extended Jaccard Graph and Stochastic Gradient Descent for Robot
This addresses robot recognition for human-robot interaction, but appears incremental as it builds on existing fusion and graph techniques.
The paper tackles robot vision recognition by proposing a feature graph fusion method that combines RGB and depth information from Kinect, using multi-Jaccard similarity and word embedding, and shows it is robust and effective on face and object datasets.
Robot vision is a fundamental device for human-robot interaction and robot complex tasks. In this paper, we use Kinect and propose a feature graph fusion (FGF) for robot recognition. Our feature fusion utilizes RGB and depth information to construct fused feature from Kinect. FGF involves multi-Jaccard similarity to compute a robust graph and utilize word embedding method to enhance the recognition results. We also collect DUT RGB-D face dataset and a benchmark datset to evaluate the effectiveness and efficiency of our method. The experimental results illustrate FGF is robust and effective to face and object datasets in robot applications.