Commit da1ba091 authored by burcharr's avatar burcharr 💬
Browse files

automatic writing commit ...

parent fe6c40df
...@@ -958,4 +958,36 @@ Subject\_term: Computational neuroscience;Computational science;Computer science ...@@ -958,4 +958,36 @@ Subject\_term: Computational neuroscience;Computational science;Computer science
Subject\_term\_id: computational-neuroscience;computational-science;computer-science;software;solar-physics}, Subject\_term\_id: computational-neuroscience;computational-science;computer-science;software;solar-physics},
keywords = {Computational neuroscience, Computational science, Computer science, Software, Solar physics}, keywords = {Computational neuroscience, Computational science, Computer science, Software, Solar physics},
file = {Full Text PDF:/home/robin/Zotero/storage/FFD42GPJ/Harris et al. - 2020 - Array programming with NumPy.pdf:application/pdf;Snapshot:/home/robin/Zotero/storage/9TW8BU97/s41586-020-2649-2.html:text/html}, file = {Full Text PDF:/home/robin/Zotero/storage/FFD42GPJ/Harris et al. - 2020 - Array programming with NumPy.pdf:application/pdf;Snapshot:/home/robin/Zotero/storage/9TW8BU97/s41586-020-2649-2.html:text/html},
}
@inproceedings{saini_human_2020,
title = {Human Activity and Gesture Recognition: A Review},
doi = {10.1109/ICONC345789.2020.9117535},
shorttitle = {Human Activity and Gesture Recognition},
abstract = {Human activity recognition is the method of extracting and predicting the movements of the human body often indoors by using any hardware device such as a camera or sensor-based device. At an earlier stage collecting data from sensors is quite expensive but recently we have smartphones and other personal devices which have accelerometer-based sensors that track our activities. {HAR} is a classification method in which people have a great interest because we can find different actions of the human body like sitting, walking, running, jumping, jogging etc. by using body-worn sensors like accelerometer, gyroscope and applying methods like convolution neural network and other deep learning methods. The main objective of this review is to study different human activities, compounds and methods that are used to recognize actions and position of the body.},
eventtitle = {2020 International Conference on Emerging Trends in Communication, Control and Computing ({ICONC}3)},
pages = {1--2},
booktitle = {2020 International Conference on Emerging Trends in Communication, Control and Computing ({ICONC}3)},
author = {Saini, Rashmi and Maan, Vinod},
date = {2020-02},
keywords = {accelerometer, deep learning, gyroscope, human activity recognition},
file = {IEEE Xplore Full Text PDF:/home/robin/Zotero/storage/UXDEKQLH/Saini and Maan - 2020 - Human Activity and Gesture Recognition A Review.pdf:application/pdf;IEEE Xplore Abstract Record:/home/robin/Zotero/storage/YZVSCT4N/9117535.html:text/html},
}
@article{wah_ng_real-time_2002,
title = {Real-time gesture recognition system and application},
volume = {20},
issn = {0262-8856},
url = {https://www.sciencedirect.com/science/article/pii/S0262885602001130},
doi = {10.1016/S0262-8856(02)00113-0},
abstract = {In this paper, we consider a vision-based system that can interpret a user's gestures in real time to manipulate windows and objects within a graphical user interface. A hand segmentation procedure first extracts binary hand blob(s) from each frame of the acquired image sequence. Fourier descriptors are used to represent the shape of the hand blobs, and are input to radial-basis function ({RBF}) network(s) for pose classification. The pose likelihood vector from the {RBF} network output is used as input to the gesture recognizer, along with motion information. Gesture recognition performances using hidden Markov models ({HMM}) and recurrent neural networks ({RNN}) were investigated. Test results showed that the continuous {HMM} yielded the best performance with gesture recognition rates of 90.2\%. Experiments with combining the continuous {HMMs} and {RNNs} revealed that a linear combination of the two classifiers improved the classification results to 91.9\%. The gesture recognition system was deployed in a prototype user interface application, and users who tested it found the gestures intuitive and the application easy to use. Real time processing rates of up to 22 frames per second were obtained.},
pages = {993--1007},
number = {13},
journaltitle = {Image and Vision Computing},
shortjournal = {Image and Vision Computing},
author = {Wah Ng, Chan and Ranganath, Surendra},
urldate = {2021-10-26},
date = {2002-12-01},
langid = {english},
keywords = {Hand segmentation, Hidden Markov models, Neural networks, Real-time gesture recognition},
} }
\ No newline at end of file
# Related Work # Related Work
Automatically detecting the current activity of a human being is a wide research field in computer science. There are many possible applications, e.g. human robot interaction, quality assessments, worker surveillance, and more. TODO: add citations for applications Automatically detecting the current activity of a human being is a wide research field in computer science. There are many possible applications, e.g. human robot interaction, quality assessments, worker surveillance, control of user interfaces and more.
## Gesture Recognition ## Gesture Recognition
In the area of gesture recognition, ... TODO, In the area of gesture recognition, we try to detect and classify specific, and closely defined gestures.
Defined gestures can be used to actively control a system. This kind of approach is not directly applicable to our task of detecting hand washing. However, it could be possible to adapt algorithms from this field to the detection of a new gesture or a new set of gestures related to hand washing. The defined gestures can e.g. be used to actively control a system @saini_human_2020. This kind of approach is not directly applicable to our task of detecting hand washing. However, it could be possible to adapt algorithms from this field to the detection of a new gesture or a new set of gestures related to hand washing.
There are camera-based approaches, that were out of scope for this work. As explained in the introduction, wrist worn devices have significant advantages over camera-based solutions that would have to be stationary, i.e. in fixed locations. There are camera-based approaches and physical measurement based approaches @saini_human_2020. The camera base approaches were out of scope for this work. As explained in the introduction, wrist worn devices have significant advantages over camera-based solutions that would have to be stationary, i.e. in fixed locations.
There also exist approaches based on inertial measurement sensors. These sensors measure movement related physical values, such as the force or acceleration, angular velocity, orientation in space. There also exist approaches based on inertial measurement sensors. These sensors measure movement related physical values, such as the force or acceleration, angular velocity, orientation in space.
The easiest detection task here is likely the one of counting steps, which can be done with worn devices. TODO citation. Gesture recognition, in general, uses similar methods as the more difficult human activity recognition @saini_human_2020.
## Human Activity Recognition ## Human Activity Recognition
\label{section:har} \label{section:har}
Recognizing more than one gesture in combination in a temporal context and deriving the current activity of the user is called human activity recognition (HAR). In this task, we want to detect a more general activity, compared to a shorter and simpler gesture. An activity can include many distinguishable gestures. However, the same activity will not always include all of the same gestures and the gestures included could be in a different order for every repetition. Activities are less repetitive than gestures, and much harder to detect in general todo citation. However, Zhu et al. have shown that the combined detection of multiple different gestures can be used in HAR tasks too @zhu_wearable_2011, which makes sense, because a human activity can consist of many gestures. Nevertheless, most methods used for HAR consist of more direct applications of machine learning to the data, without the detour of detecting specific gestures contained in the execution of an activity. Recognizing more than one gesture or body movement in combination in a temporal context and deriving the current activity of the user is called human activity recognition (HAR). In this task, we want to detect a more general activity, compared to a shorter and simpler gesture. An activity can include many distinguishable gestures. However, the same activity will not always include all of the same gestures and the gestures included could be in a different order for every repetition. Activities are less repetitive than gestures, and harder to detect in general @zhu_wearable_2011. However, Zhu et al. have shown that the combined detection of multiple different gestures can be used in HAR tasks too @zhu_wearable_2011, which makes sense, because a human activity can consist of many gestures. Nevertheless, most methods used for HAR consist of more direct applications of machine learning to the data, without the detour of detecting specific gestures contained in the execution of an activity.
Methods used in HAR include classical machine learning methods as well as deep learning @liu_overview_2021 @bulling_tutorial_2014. The classical machine learning methods rely on features of the data obtained by feature engineering. These methods include but are not limited to Random Forests, Hidden Markov Models (HMM), Support Vector Machines (SVM), and more. The features can frequency-domain based and time-domain based, but usually both are used at the same time to train these conventional models @liu_overview_2021. Methods used in HAR include classical machine learning methods as well as deep learning @liu_overview_2021 @bulling_tutorial_2014. The classical machine learning methods rely on features of the data obtained by feature engineering. These methods include but are not limited to Random Forests, Hidden Markov Models (HMM), Support Vector Machines (SVM) $k$-nearest neighbors algorithm, and more. The features can frequency-domain based and time-domain based, but usually both are used at the same time to train these conventional models @liu_overview_2021.
#### Deep neural networks #### Deep neural networks
Recently, deep neural networks have taken over the role of the state of the art machine learning method in the area of human activity recognition @bock_improving_2021, @liu_overview_2021. Deep neural networks are universal function approximators @bishop_pattern_2006, and are known for being easy to use on "raw" data. They are "artificial neural networks" consisting of multiple layers, where each layer contains a set amount of nodes that are connected to the nodes of the following layer. Simple neural networks where all nodes of a layer are connected to all nodes in the following layer are often called "fully connected neural networks (FC-NN or FC)". Recently, deep neural networks have taken over the role of the state of the art machine learning method in the area of human activity recognition @bock_improving_2021, @liu_overview_2021. Deep neural networks are universal function approximators @bishop_pattern_2006, and are known for being easy to use on "raw" data. They are "artificial neural networks" consisting of multiple layers, where each layer contains a set amount of nodes that are connected to the nodes of the following layer. Simple neural networks where all nodes of a layer are connected to all nodes in the following layer are often called "fully connected neural networks (FC-NN or FC)".
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment