Facebook posted in a blog post about their new large-scale machine learning which was hailed Rosetta that reads the content of the text in a picture or video.
Rosetta was made for preventing hate speech and offensive comments. The new tool will easily identify the content of the shared post if it is violating the app’s hate-speech policy. The new tool recognized different language because of the global platform of Facebook. The app will detect if the image violated any policy and it will be recognized the text so that it can be translated according to the specific region. Based on Faster R-CNN, Facebook adopted it for text detection, which is a state-of-the-art object detection network.
Facebook added “Moreover, it’s difficult to get human-annotated data for many of the languages. Although SynthText has been helpful as a way to bootstrap training, it’s not yet a replacement for human-annotated data sets. We are therefore exploring ways to bridge the domain gap between our synthetic engine and real-world distribution of text on images.”