Real-time static custom gestures recognition based on skeleton hand

Osipov, Alexander; Ostanin, Mikhail

doi:10.1109/nir52917.2021.9665809

Cited by 5 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Experiment is repeated 20 times for each hand gesture and specified distance. Table (4) shows that at 1 m, the accuracy reaches 100% for all hand gestures. This means that at this distance, all hand gestures can be correctly classified.…”

Section: Test the Proposed Model When Static Image Mode Is Undefault ...mentioning

confidence: 97%

“…In MP, Hand Gestures were detected by two faces. The first was the Hand Detection via palms detector and the second was the Hand Landmark via 21 three dimensional points on every part of the hand gesture as shown in the Figure (1) [4]. Subhangi Adhikary et.al [15] saved the new data which contains 21 points on every part of the Hand Gesture in a file called CSV, and this data were used to train and test the system.…”

Section: Related Workmentioning

confidence: 99%

“…Figure (7) shows 21 points 3D on each part of the palm. To classify hand gestures, the value of coordinates of the points (4,8,12,16,20) are compared. These points represent the fingertips with the value of coordinates of the points (3, 5, 9, 13, 17) that represent the middle points relative to the x and y axes.…”

Section: Gesture Recognizer Modelmentioning

confidence: 99%

“…Computer vision is the only technique that can extract information from incoming images [3]. The MediaPipe algorithm is used to detect and classify the hand gestures [4]. The goal in this research is to detect hand gestures using the MP algorithm via the camera sensor, classify hand gestures, and transmit the instructions wirelessly.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Tracked Robot Control with Hand Gesture Based on MediaPipe

Wameed,

M. ALKAMACHI,

Erçelebi

2023

alkej

View full text Add to dashboard Cite

Hand gestures are currently considered one of the most accurate ways to communicate in many applications, such as sign language, controlling robots, the virtual world, smart homes, and the field of video games. Several techniques are used to detect and classify hand gestures, for instance using gloves that contain several sensors or depending on computer vision. In this work, computer vision is utilized instead of using gloves to control the robot's movement. That is because gloves need complicated electrical connections that limit user mobility, sensors may be costly to replace, and gloves can spread skin illnesses between users. Based on computer vision, the MediaPipe (MP) method is used. This method is a modern method that is discovered by Google. This method is described by detecting and classifying hand gestures by identifying 21 three-dimensional points on the hand, and by comparing the dimensions of those points. This is how the hand gestures are classified. After detecting and classifying the hand gestures, the system controls the tracked robot through hand gestures in real time, as each hand gesture has a specific movement that the tracked robot performs. In this work, some important paragraphs concluded that the MP method is more accurate and faster in response than the Deep Learning (DL) method, specifically the Convolution Neural Network (CNN). The experimental results shows the accuracy of this method in real time through the effect of environmental elements decreases in some cases when environmental factors change. Environmental elements are such light intensity, distance, and tilt angle (between the hand gesture and camera).The reason for this is that in some cases, the fingers are closed together, and some fingers are not fully closed or opened and the accuracy of the camera used is not good with the changing environmental factors. This leads to the inability of the algorithm used to classify hand gestures correctly (the classification accuracy decrease), and thus response time of the tracked robot's movement increases. That does not present possibility for the system to determine whether the finger is closed or opened.

show abstract

Section: Test the Proposed Model When Static Image Mode Is Undefault ...mentioning

confidence: 97%

Section: Related Workmentioning

confidence: 99%

Section: Gesture Recognizer Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Tracked Robot Control with Hand Gesture Based on MediaPipe

Wameed,

M. ALKAMACHI,

Erçelebi

2023

alkej

View full text Add to dashboard Cite

show abstract

“…The data of the extracted coordinates were stored in pickle format for further processing. The pickle format is a convenient format for storing and loading complex data structures in Python and is commonly used in machine learning applications [1].…”

Section: Data Collectionmentioning

confidence: 99%

Sign Language Based Video Calling App

Shaikh¹,

Hadiyal²,

Choudhary³

et al. 2023

IJSRCSEIT

View full text Add to dashboard Cite

Deaf and hard of hearing people communicate with others and within their own communities by using sign language. Computer recognition of sign language begins with the learning of sign gestures and continues through the production of text and speech. There are two types of sign gestures: static and dynamic. Both gesture recognition systems are crucial to the human community, even if static gesture recognition is easier than dynamic gesture recognition. We have conducted research on the steps required to convert static American sign language (ASL) to readable text and selected the best available methods to do so. Examined general steps are the data collection, pre-processing, transformation, feature extraction, and classification. There are also some recommendations for further study in this field.

show abstract