Significant improvement for temporal consistency in video semantic segmentation

Semantic segmentation is a far tricker task for video than for static images, either resulting in temporally inconsistent – or costly and inaccurate – predictions. Momentum Adapt is an unsupervised online method that improves temporal performance to deliver the consistency your AI applications need. Uncover how this approach outperforms state-of-the-art algorithms in adapting to even the most severe environmental changes.

Find out more about this novel approach to improving semantic segmentation performance in the whitepaper by Amirhossein Hassankhani, Hamed Rezazadegan Tavakoli and Esa Rahtu.
https://papers.bmvc2023.org/0709.pdf

Source: Nokia YouTube