Digital image processing and computer vision techniques are nowadays used for a wide variety of applications. With the rapid proliferation of inexpensive acquisition and storage devices, multimedia objects can be easily created, stored, transmitted, modified and tampered with by anyone.
Digital Image Processing
Within this context, we developed an expertise on a series of different topics:
- Image and video analysis (e.g. multispectral imaging for food quality assessment and other industrial applications, object recognition in low-power platforms, etc.).
- 3D processing (e.g. scene reconstruction and analysis for automatic docking assistance, posture monitoring for clinical analysis, etc.).
- Video and image coding (e.g. exploiting distributed processing environments, coding of image descriptors, etc.).
Multimedia Forensics
During its lifetime, a digital object might go through several processing stages, including multiple analog-to-digital (A/D) and digital-to-analog (D/A) conversions, coding and decoding, transmission, and editing.
Traces left by processing operations can be exploited to develop solutions that enable to expose the past history of multimedia content. Leveraging these footprints, we develop techniques with different goals:
- Detect forgeries in images and videos.
- Link an image or video to the originating device used to shoot it.
- Discriminate between original imagery and fakes obtained through CNNs and GANs.
- Analyze satellite images to detect manipulation.
- Expose forgeries on images belonging to scientific publications to fight author misconduct.