Research topics

Computer vision

  • AI and machine learning for computer vision

  • Visual attention modelling

  • Hand gesture recognition

  • Scalable and robust visual descriptors

  • Compression-domain feature extraction – HEVC and JPEG2000 coded bitstream analysis

  • Motion modelling – phase features

  • Multimodal data fusion

  • Texture modelling

  • Video and image analysis for surveillance, security, creative industries and healthcare applications

Content forensics and security

  • AI and machine learning for content forensics and security

  • Spread sectrum image/video watermarking

  • Robustness to content adaptation attacks

  • Watermrking for HEVC video

  • Content-based fingerprinting

  • Fake visual content detection

Signal processing

  • Compressive sensing – fast and high efficient reconstruction

  • Graph signal processing

  • Spatially adaptive wavelets with no side information

  • Adaptive mother wavelet packet models for texture descriptions

  • Integer-to-integer mapping for DCT, DST and WHT

Video and image coding

  • Video coding standards

  • Ultra High Definition (UHD) video coding

  • High dynamic range (HDR) image and video coding

  • Multiple description video coding

  • Compressive sensing application on image and video representation

  • Scalable coding and content adaptation

  • Lossless and near-lossless coding