Spatiotemporal features have significant importance in human action recognition, as they provide the actor's shape and motion characteristics specific to each action class. This paper presents a new deep spatiotemporal human action representation, \Deep Temporal Motion Descriptor (DTMD)", which shares the attributes of holistic and deep learned features. To generate the DTMD descriptor, the actor's silhouettes are gathered into single motion templates through applying motion history images. These motion templates capture the spatiotemporal movements of the actor and compactly represents the human actions using a single 2D template. Then, deep convolutional neural networks are used to compute discriminative deep features from motion history templates to produce DTMD. Later, DTMD is used for learn a model to recognise human actions using a softmax classifier. The advantage of DTMD comes from (i) DTMD is automatically learned from videos and contains higher dimensional discriminative spatiotemporal representation as compared to handcrafted features; (ii) DTMD reduces the computational complexity of human activity recognition as all the video frames are compactly represented as a single motion template; (iii) DTMD works e ectively for single and multiview action recognition. We conducted experiments on three challenging datasets: MuHAVI-Uncut, iXMAS, and IAVID-1. The experimental findings reveal that DTMD outperforms previous methods and achieves the highest action prediction rate on the MuHAVI-Uncut dataset
Deep temporal motion descriptor (DTMD) for human action recognition
2020-08-05
AR/0000024013
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
DDC: | 629 |
Action recognition using edge trajectories and motion acceleration descriptor
British Library Online Contents | 2016
|Kinematic Spline Curves: A temporal invariant descriptor for fast action recognition
British Library Online Contents | 2018
|Action recognition using 3D DAISY descriptor
British Library Online Contents | 2014
|Histogram of oriented rectangles: A new pose descriptor for human action recognition
British Library Online Contents | 2009
|Histogram of 3D Facets: A depth descriptor for human action and hand gesture recognition
British Library Online Contents | 2015
|