This paper presents a hierarchical clustering algorithm that uses multiple heterogeneous representations to group trajectories for video event detection, enabling it to effectively identify both dominant and rare events and achieve better performance than state-of-the-art methods.
Oct 7, 2018
This paper proposes a novel event-based video synopsis method that addresses object occlusion by using trajectory kinematics descriptors and affinity propagation to cluster similar kinematic events, resulting in clearer and more efficient summary videos compared to state-of-the-art techniques.
May 8, 2017
This paper introduces a new temporal coherent face descriptor for video gender recognition, which constructs a unified feature description from continuous video frames to eliminate the need for face normalization and alignment, achieving real-time, state-of-the-art results using a support vector machine classifier.
Jun 1, 2015
To overcome the problem of comparing object trajectories with varying lengths, this paper proposes a novel trajectory kinematics descriptor using Frenet-Serret frames, which achieves superior F-measure scores in event clustering compared to existing state-of-the-art methods.
May 24, 2015