Department of Computational Perception
Department of
Computational Perception
Johannes Kepler Universit+AOQ-t Linz

Home  –  Mission  –  Teaching  –  People  –  Research  –  Media  –  Impressum


Project Title: SocialFilm

Sponsor:  Austrian Research Promotion Agency (FFG), Innovation Voucher

Industrial Partner: emotion3D

Duration: 5 months (July 2017 November 2017)

Persons involved:

Markus Schedl (Principal Investigator)
Bruce Ferwerda (PostDoc researcher)


Current smart phones and digital cameras allow users to create high-quality images easily without the need for professional expertise, e.g., in the field of photography. Based on fully automatic image analysis (for example, detection of closed eyes, automatic white balance, etc.), the device supports the user in taking pictues. In contrast, the creation of good short videos with the smart phone is still not easy to accomplish for the amateur users, since good camera movement puts significantly higher demands on them. To overcome this issue, emotion3D is working on a fully automated software tool to support users in the video production process on smart phones. This tool analyzes the quality of the video during recording, using intelligent image analysis (such as Optical Flow or Visual Odometry) and IMU sensors. It can identify and alert the user about aesthetic mistakes, e.g., poor camera direction or scene composition.

This project aims at analyzing possibilities to personalize the video recording process on smart phones, by addressing the following tasks:
  1. defining possibilities to personalize video content and recording applications
  2. identifying criteria to measure popularity, impact, etc. of video content on social media platforms, such as Instagram
  3. reviewing filtering and storyboard functions offered by social media platforms, with a special focus on Instagram, and creating a ranking of these functions with respect to the criteria identified in the previous task
  4. analyzing possibilities to create personalized storyboards, based on social media content of users or groups of users, as well as extracting "stories" from several Instragram images
  5. identifying research questions from the results of all above task, which can serve as basis for further collaborations

last edited by ms on 2017-11-08