By Paula Viana (INESC)
The digital revolution is one thing that is not being slowed down by COVID-19 — indeed, it is speeding up! FotoinMotion is part of that revolution, and is proud to present its photo-to-video solution to the worldwide multimedia community at the 28th ACM International Conference on Multimedia.
ACM Multimedia, taking place in Seattle and virtually, is the premier international conference in multimedia, covering multiple emerging fields focusing on advancing the research and applications of many forms of media, including but not limited to images, text, audio, speech, music, sensor and social data.
Our paper “Semantic Storytelling Automation: A Context-Aware and Metadata-Driven Approach” will be discussed in a Q&A session on the October 13th, 18:00-19:00 (UTC), and on the October 14th October, 4:00-5:00 (UTC).
The paper describes the FotoInMotion framework that enables transforming a still photograph into a semantic-aware video clip. Sensing the capturing context and analyzing the image using computer vision and machine learning techniques enables generating valuable information to be used in the creative process. This metadata is used to automatically create an engaging video with several immersive effects that is able to convey the semantics in the content, the environment and the application scenario.
Stay tuned to the ACM MM website, watch our demo video and join us in the Q&A sessions!