2022
DOI: 10.48550/arxiv.2202.13096
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Continuous Human Action Recognition for Human-Machine Interaction: A Review

Abstract: With advances in data-driven machine learning research, a wide variety of prediction models have been proposed to capture spatio-temporal features for the analysis of video streams. Recognising actions and detecting action transitions within an input video are challenging but necessary tasks for applications that require real-time human-machine interaction. By reviewing a large body of recent related work in the literature, we thoroughly analyse, explain and compare action segmentation methods and provide deta… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
1
1

Relationship

1
1

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 153 publications
(272 reference statements)
0
2
0
Order By: Relevance
“…Testing on a single Nvidia Geforce RTX 2080Ti and 4 CPU cores, the pipelines takes about 175ms to process a 50 frames of data. Although our results are promising, additional optimization and model pruning techniques are required for the action recognition model [13], such as optimizing the computation graph generated in Pytorch with TensorRT, to deploy our pipeline onto an edge-AI system that can be built around, for example a single Jetson AGX Xavier.…”
Section: Action Recognition Models Evaluationmentioning
confidence: 99%
“…Testing on a single Nvidia Geforce RTX 2080Ti and 4 CPU cores, the pipelines takes about 175ms to process a 50 frames of data. Although our results are promising, additional optimization and model pruning techniques are required for the action recognition model [13], such as optimizing the computation graph generated in Pytorch with TensorRT, to deploy our pipeline onto an edge-AI system that can be built around, for example a single Jetson AGX Xavier.…”
Section: Action Recognition Models Evaluationmentioning
confidence: 99%
“…Even if this can be considered a simple task, it has puzzled computer vision scholars for several decades [2][3][4]. Throughout this period, human action recognition has been widely adopted by various scientific fields, such as human-machine interaction [5,6], medical assistive technologies [7][8][9][10], surveillance systems [11,12], sports analysis [13], and human-robot interaction [14,15]. Similarly, it assists in path planning for tasks like social collision avoidance and route optimization [16] in autonomous navigation [17].…”
Section: Introductionmentioning
confidence: 99%