For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Recent Advances in Video Action Recognition with 3D Convolutions
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences
Publication Date: 2021/06/01
Online ISSN: 1745-1337
Type of Manuscript: INVITED PAPER (Special Section on Image Media Quality)
video recognition, action recognition, 3D convolutions, survey,
Full Text: FreePDF
The performance of video action recognition has improved significantly in recent decades. Current recognition approaches mainly utilize convolutional neural networks to acquire video feature representations. In addition to the spatial information of video frames, temporal information such as motions and changes is important for recognizing videos. Therefore, the use of convolutions in a spatiotemporal three-dimensional (3D) space for representing spatiotemporal features has garnered significant attention. Herein, we introduce recent advances in 3D convolutions for video action recognition.