ShowmaxLab

About

ShowmaxLab is a joint research space between Showmax and FIT CTU for collaboration on machine learning and computer vision. The main focus is at improving understanding of video and recommendations for the Showmax users.

Selected projects

Current

Action detection in videos

The aim of this project is to recognize what people are doing in videos. As recognizing human actions is a complex high-level task, very deep convolutional networks trained on large datasets are used.

Researcher: Ondřej Bíža

research proposal

Past

Detecting scene locations

In this project, techniques for understanding where movie scenes take place were explored. The final system used convolutional neural networks trained on the Imagenet dataset and fine-tuned on our custom dataset.

Researcher: Lukáš Lopatovský

presentation?

Understanding movie audio

While audio understanding has not enjoyed the same popularity as image or video recognition, there is a lot of useful information that can be extracted from movie audio. In this project, a neural network trained on a large corpus of sound was used to detect and analyze individual movie scenes. The system was particularly good at recognizing dialogues and action scenes.

Researcher: Ondřej Bíža

presentation

Contact

Showmax Engineering

This email address is being protected from spambots. You need JavaScript enabled to view it.

Follow Us

Copyright (c) Data Science Laboratory @ FIT CTU 2014–2016. All rights reserved.