K-HOG Unsupervised Keyframe Identifier (K-HUKI): Extracting action-rich frames with HOG Features and Unsupervised Learning

This Report proposes a pioneering method for keyframe identification in activity interpretation. It seamlessly integrates HOG features for informative representation and the flexibility of unsupervised learning through K-means clustering. This approach, named K-HUKI, is further empowered by a robust architecture combining 3D-CNNs for capturing temporal and spatial information and GRUs for handling perpetual dependencies. Extensive evaluation on the UCF101 dataset demonstrates a remarkable improvement in accuracy. By identifying action-rich frames at exceptional speed (almost 5-fold) compared to CNN based approach's and achieving unparalleled accuracy, K-HUKI establishes itself as a groundbreaking advancement in keyframe detection.