Innovative signal processing for multimedia

Description

Many traditional image and video processing tasks have been revolutionized by the advent of artificial intelligence, paving the way to several promising research directions. Deep neural networks can provide state-of-the-art solutions to inverse problems like deconvolution, denoising, image restoration, image reconstruction, and image super-resolution. Deep architectures like autoencoders can be applied to image and video compression, with unprecedented results in terms of rate-distortion performance. Advanced generative models can be used to produce very realistic synthetic media, which can be used to enable digital twins or enhance existing datasets. On the other hand, the facility with which images and videos can be altered or even synthetically generated by AI stimulates research on innovative forensic tools for verifying the origin and integrity of multimedia.

ERC sectors 

  • PE7_7 Signal processing
  • PE6_2 Distributed systems, parallel computing, sensor networks, cyber-physical systems
  • PE6_5 Security, privacy, cryptology, quantum cryptography
  • PE6_7 Artificial intelligence, intelligent systems, natural language processing
  • PE6_8 Computer graphics, computer vision, multimedia, computer games
  • PE6_11 Machine learning, statistical data processing and applications using signal processing (e.g. speech, image, video)

Keywords 

  • Deep neural networks
  • Autoencoders
  • Transformers
  • Generative models
  • Inverse problems
  • Multimedia forensics