Open-vocab Temporal Action Detection

Open-vocab Temporal Action Detection is a computer vision technique aimed at detecting and recognizing unseen action categories from videos by learning action features in temporal sequences, enabling accurate identification and localization of newly emerged or rare actions. This technology has a wide range of applications, including intelligent surveillance, content recommendation, and human-computer interaction.