Doctoral Schools WUT

Search Engine for Promoters and Research Areas

Wykaz obszarów badawczych związanych z tagiem Kodowanie:

# Obszar badawczy Dziedzina naukowa
1 New video compression methods based on neural networks: The research problem is devoted to the development of new architectures of convolutional neural networks and other video signal processing components for effective compression of video sequences. These other elements include the use of orthogonal transformations of the DCT and DWT type, descriptors allowing for the formation of the context. Basically, the network architectures for compression are based on the autoencoder structure. However, the relationship between the layers and the structure of the layers itself depends on many hyper parameters. The goal is to find a solution that will give the best possible compression efficiency at a limited computational cost. It is advisable to use learning in the GAN configuration to adjust the compression to the human vision system.
2 New video compression methods for low-latency transmission: The research issue is devoted to the development of new algorithms for selecting the encoding modes and adjusting the compression rate for video sequences. The aim of the work is to develop software methods for creating information streams intended for signal transmission in real-time conditions with low delay (several milliseconds). Traditional video compression schemes use the so-called Intra keyframes encoded entirely as single images without reference to previous images and allowing playback to begin without reference to the history. The cost of this functionality is a strongly and temporarily increased bitstream. Variations in the complexity of the encoded video content also vary the instantaneous stream bit rate. Through the efficient algorithm of compression rate adjustment and Intra / Inter mode selection for blocks within frames, bit rate balance and full re-synchronization can be achieved after decoding a few frames.