Computer Vision
Projects
MORPHCAST
Real time video creation according to your emotions
2018-2020
funded by: Regione Toscana | CYNNY S.P.A.
MORPHCAST is an innovative application made by the innovative SME CYNNY S.p.a. that personnalises video contents in real time leveraging the emotional state of the viewer. The system is able to run on both mobile and desktop platforms through a web browser. The profiling is done using information obtained only from the user’s face, such as age, gender, expressed emotions, arousal, valence, head position and 30+ features. To obtain such information, a face analysis system was implemented, together with a “full stack” of computer vision and deep learning algorithms to extract the user pose and estimate demographic information. The aim of this project is to optimize the stack to run in javascript inside of a browser.
IDEHA
Innovation for Data Elaboration in Heritage and Arts
Nov 2018 – Apr 2021
Funded by: MIUR – PON | CNR
IDEHA will create an open IT platform for Cultural Heritage, combining both digital content from traditional repositories and information generated in real time by users or environmental sensors; aggregating, processing and understanding the data by utilizing new learning technologies to build services that can be used by different users profiled through specific multimodal applications.
Predicting Action Progress
Action understanding and prediction for humans assistance
2018-2019
In this project we address the problem of predicting action progress in videos. This is an extremely important task because, on the one hand, it can be valuable for a wide range of applications and, on the other hand, it facilitates better action detection results.
To solve this problem we introduce a novel approach, named ProgressNet, capable of predicting when an action takes place in a video, where it is located within the frames, and how far it has progressed during its execution.
GANIMEDE
Object detection and recognition
2018-2019
funded by: Leonardo – Finmeccanica
GANIMEDE is a technology transfer project in collaboration with the Italian multinational aerospace, defense and security company Leonardo. In this project MICC is developing a video analysis framework to detect and recognise objects of security interest.