Ubiquitous and affordable digital cameras have enabled users to take pictures and videos everywhere and anytime. This has led to an explosion of the amount of image material both amateurs and professionals have to work with. Assessing, selecting, editing, organizing, and annotating this large amount of visual data is a tedious and time-consuming process, as it involves a lot of manual labor with only minimal basic computational support available to users.
This project aims to address major gaps and challenges in automating photowork, i.e. the processes of picture assessment, enhancement, editing, and annotation, with a particular focus on large content collections. These collections may be composed of personal or professional photos and videos (e.g. vacation photos, wedding photo shoot), possibly from multiple users and/or devices, by images shared in social networks and online sites, and third-party repositories.
Human Sixth Sense Framework
For telepresence applications to create the illusion of being in the same place with the other participants, visual characteristics are an important component. Visuals of each person need to be of high photographic and picture quality. Any enhanced views, such as stitching participants together for an increased sense of presence, require quality assessment, photometric matching and rendering capabilities. User preferences important to make virtual meetings more personalized and interactive. Many of these issues can be addressed by applying the above methods for pictures to live video streams.
- What criteria do people use for photowork?
- How can photographic and semantic characteristics be combined for picture assessment and selection?
- How can picture comparisons be made easier?
- How can pictures be enhanced through intelligent segmentation and combination of regions?
- How can pictures be organized semi-automatically with satisfying results?
- How can the experience of browsing a picture collection be enhanced?
- How can people's individual preferences and tastes be modeled?
- How can computer interfaces for photowork be made more intuitive?
We have established the following collaborations for this project:
- Philip Fu, NTU (human-computer interaction)
- Shih-Cheng Yen, NUS (eye-tracking)
- Fook Kee Chua, NUS (eye-tracking)
- Sabine Süsstrunk, EPFL (skin color and face detection)
Please refer to our publications.