Fast Dynamic Convolutional Neural Networks for Visual Tracking
share
Most of the existing tracking methods based on CNN(convolutional neural networks) are too slow for real-time application despite the excellent tracking accuracy compared with the traditional ones. In this paper, a fast dynamic visual tracking algorithm combining CNN based MDNet(Multi-Domain Network) and RoIAlign was developed. The major problem of MDNet also lies in the time efficiency. Considering the computational complexity of MDNet is mainly caused by the large amount of convolution operations, a RoIPool layer which could conduct the convolution over the whole image instead of each RoI was added to accelerate the calculation. With RoIPool employed, the computation speed has been increased but the tracking accuracy has dropped simultaneously. RoIPool could lose some positioning accuracy because it can not handle locations represented by floating numbers. So RoIAlign, instead of RoIPool, which can process floating numbers of locations by bilinear interpolation has been added to the network. The results show the target localization accuracy has been improved and it hardly increases the computational cost. These strategies can accelerate the processing and make it 5x faster than MDNet with very low impact on accuracy. The proposed algorithm has been evaluated on two benchmarks: OTB100 and VOT2016, on which high accuracy and speed have been obtained. The influence of the network structure and training data are also discussed with experiments.
READ FULL TEXT