Abstract: Recent advancements in convolutional neural network (CNN)-based techniques for remote sensing pansharpening have markedly enhanced image quality. However, conventional convolutional modules ...
Abstract: The aim of scene semantic segmentation is to label each pixel with a class which it belongs to in high level cognition. State-of-art works mainly adapt convolutional neural networks ...
This project is based on block wise frequency domain convolution of input audio and Head Transfer Function (HRTF) database, using FFT based Overlap Add method to achieve real-time binaural spatialized ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results