Real-Time MDNet
This provides a faster and more accurate visual tracking solution for applications like surveillance and robotics, but it is incremental as it builds upon the existing MDNet framework.
The paper tackles the problem of slow visual tracking by accelerating the MDNet algorithm, achieving approximately 25 times speed-up with almost identical accuracy and outperforming state-of-the-art real-time methods on multiple benchmarks.
We present a fast and accurate visual tracking algorithm based on the multi-domain convolutional neural network (MDNet). The proposed approach accelerates feature extraction procedure and learns more discriminative models for instance classification; it enhances representation quality of target and background by maintaining a high resolution feature map with a large receptive field per activation. We also introduce a novel loss term to differentiate foreground instances across multiple domains and learn a more discriminative embedding of target objects with similar semantics. The proposed techniques are integrated into the pipeline of a well known CNN-based visual tracking algorithm, MDNet. We accomplish approximately 25 times speed-up with almost identical accuracy compared to MDNet. Our algorithm is evaluated in multiple popular tracking benchmark datasets including OTB2015, UAV123, and TempleColor, and outperforms the state-of-the-art real-time tracking methods consistently even without dataset-specific parameter tuning.