no code implementations • 11 Mar 2024 • Tunc Alkanat, Erkut Akdag, Egor Bondarev, Peter H. N. de With
Temporal localization of driving actions plays a crucial role in advanced driver-assistance systems and naturalistic driving studies.
no code implementations • 11 Mar 2024 • Erkut Akdag, Zeqi Zhu, Egor Bondarev, Peter H. N. de With
The model uses 2D-pose features as the positional embedding of the transformer architecture and spatio-temporal features as the main input to the encoder of the transformer.
no code implementations • 11 Mar 2024 • Ivo P. C. Kersten, Erkut Akdag, Egor Bondarev, Peter H. N. de With
Second, we compare the performance of different feature extractors for our anomaly detection method on the UCF-Crime and Throwing-Action datasets.