Learning-Based Video Super-Resolution Reconstruction Using Particle Swarm Optimization

In this study, a learning-based video super-resolution (SR) reconstruction approach using particle swarm optimization (PSO) is proposed. First, a 5 x 5 x 5 motion-compensated volume containing five 5 x 5 motion-compensated patches is extracted and the orientation of the volume is determined for each pixel in the "central" reference low-resolution (LR) video frame. Then, the pixel values of the "central" reference high-resolution (HR) video frame are reconstructed by using the corresponding SR reconstruction filtering masks, based on the orientation of the volume and the coordinates of the pixels to be reconstructed. To simplify the PSO learning processes for determining the weights in SR reconstruction filtering masks, simple mask flippings are employed. Based on the experimental results obtained in this study, the SR reconstruction results of the proposed approach are better than those of three comparison approaches, whereas the computational complexity of the proposed approach is higher than those of two "simple" comparison approaches, NN and Bicubic, and lower than that of the recent comparison approach, modified NLM.