The basis for training the model was “Thesaurus Linguae Aegyptiae”, the largest database of ancient Egyptian texts with translations. The accuracy of the system was checked both using automated methods and with the participation of professional Egyptologists from the Higher School of Economics of the National Research University.
Developed technology could make learning the ancient Egyptian language accessible even to beginners. It will also be possible to use it in mobile applications in the future. For example, a tourist will be able to point his phone’s camera at an artifact in the museum and instantly see the translation of the inscription. The system could also make it easier to decipher new texts and aid the study of Egyptology students.
One of the challenges of creating the model was the sheer number of characters – unlike alphabetic systems, there are hundreds of characters in the ancient Egyptian letter, many of which are similar to each other. For recognition, OCR models were used that take into account the context and make it possible to more accurately distinguish visually similar hieroglyphs.
To increase the amount of data, researchers used propagation models that produce realistic images of texts. The average row in the database contains approximately 35 characters.
Currently, access to the system can be provided upon request by experts. The developers presented the results of their work at SIGGRAPH, ICCV, IEEE ISMAR and ACM Multimedia international conferences.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
