Deep convolutional correlation iterative particle filter for visual tracking

This work proposes a novel framework for visual tracking based on the integration of an iterative particle filter, deep convolutional neural network, and correlation filter. The filter enables particles to correct themselves converge target position. We employ strategy assess likelihood after iterations by applying K-means clustering. Our approach ensures consistent support posterior distribution. Thus, we do not need perform resampling at every video frame, improving utilization prior distribution information. Experimental results two different benchmark datasets show that our tracker performs favorably against state-of-the-art methods.