Paper Info
Title
Salient Superpixel Visual Tracking with Coarse-to-Fine Segmentation and Manifold Ranking.
Abstract
We propose a novel salient superpixel based tracking algorithm using Coarse-to-Fine segmentation on graph model, where target state is estimated by a combination of pixel-level cues and middle-level cues to achieve accurate target appearance model. We exploit temporal optical flow and color distribution characteristics as coarse grained information from pixel-level processing, and propagate to fine-grained superpixels to improve initial target appearance segmentation from bounding box annotations. Our algorithm constructs a graph model with manifold ranking by improved superpixels to estimate the saliency of target foreground and background in subsequent frames. The tracking result is located by calculating the weight of multi-scale box, where the weight depends on the similarity of scores of foreground and background superpixels in the scale box. We compared our algorithm with the existing techniques in OTB100 dataset, and achieved substantially better performance.
Year
Venue
Field
2018
BICS
Pattern recognition,Salience (neuroscience),Segmentation,Computer science,Active appearance model,Eye tracking,Artificial intelligence,Manifold ranking,Optical flow,Salient,Minimum bounding box
DocType
Citations 
PageRank 
Conference
0
0.34
References 
Authors
18
2
Name
Order
Citations
PageRank
Jin Zhan100.68
Huimin Zhao220623.43