Full Text Views
71
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.
Kensho HARA
the National Instutite of Advanced Industrial Science and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Kensho HARA, "Recent Advances in Video Action Recognition with 3D Convolutions" in IEICE TRANSACTIONS on Fundamentals,
vol. E104-A, no. 6, pp. 846-856, June 2021, doi: 10.1587/transfun.2020IMP0012.
Abstract: 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.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2020IMP0012/_p
Copy
@ARTICLE{e104-a_6_846,
author={Kensho HARA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Recent Advances in Video Action Recognition with 3D Convolutions},
year={2021},
volume={E104-A},
number={6},
pages={846-856},
abstract={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.},
keywords={},
doi={10.1587/transfun.2020IMP0012},
ISSN={1745-1337},
month={June},}
Copy
TY - JOUR
TI - Recent Advances in Video Action Recognition with 3D Convolutions
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 846
EP - 856
AU - Kensho HARA
PY - 2021
DO - 10.1587/transfun.2020IMP0012
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E104-A
IS - 6
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - June 2021
AB - 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.
ER -