CC BY 4.0Zhong, ZeyunZeyunZhong2023-07-192023-07-192023https://publica.fraunhofer.de/handle/publica/445798https://doi.org/10.24406/publica-165410.24406/publica-1654The ability to anticipate possible human actions in the distant future is of fundamental interest for a wide range of applications, including autonomous driving, surveillance, and human-robot interaction. Consequently, various methods have been presented for action anticipation in recent years, with deep learning-based approaches being particularly popular. In this work, we give a short overview of the recent advances of long-term action anticipation algorithmsenconference paper