Duodepth: Static Gesture Recognition Via Dual Depth Sensors

Chugunov, Ilya; Zakhor, Avideh

doi:10.1109/icip.2019.8803665

Cited by 2 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors reported state-of-the-art results using ModelNet40 shape classification benchmark. Further architectural extensions include PointNet++ [23], Temporal PointNet [24] for one-shot learning, and Dual PointNet for fusing information from two depth cameras [25].…”

Section: Introductionmentioning

confidence: 99%

A PointNet-Based Solution for 3D Hand Gesture Recognition

Mirsu

Simion

Căleanu

et al. 2020

Sensors

View full text Add to dashboard Cite

Gesture recognition is an intensively researched area for several reasons. One of the most important reasons is because of this technology’s numerous application in various domains (e.g., robotics, games, medicine, automotive, etc.) Additionally, the introduction of three-dimensional (3D) image acquisition techniques (e.g., stereovision, projected-light, time-of-flight, etc.) overcomes the limitations of traditional two-dimensional (2D) approaches. Combined with the larger availability of 3D sensors (e.g., Microsoft Kinect, Intel RealSense, photonic mixer device (PMD), CamCube, etc.), recent interest in this domain has sparked. Moreover, in many computer vision tasks, the traditional statistic top approaches were outperformed by deep neural network-based solutions. In view of these considerations, we proposed a deep neural network solution by employing PointNet architecture for the problem of hand gesture recognition using depth data produced by a time of flight (ToF) sensor. We created a custom hand gesture dataset, then proposed a multistage hand segmentation by designing filtering, clustering, and finding the hand in the volume of interest and hand-forearm segmentation. For comparison purpose, two equivalent datasets were tested: a 3D point cloud dataset and a 2D image dataset, both obtained from the same stream. Besides the advantages of the 3D technology, the accuracy of the 3D method using PointNet is proven to outperform the 2D method in all circumstances, even the 2D method that employs a deep neural network.

show abstract

Section: Introductionmentioning

confidence: 99%

A PointNet-Based Solution for 3D Hand Gesture Recognition

Mirsu

Simion

Căleanu

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

Hand Signs Recognition from Cellphone Camera Captured Images for Deaf-Mute Persons

Masum

Mollah

2021

Artificial Intelligence and Technologies

View full text Add to dashboard Cite

Communication has been a vital point of human civilization and technological advancement. Hand gesture recognition is a form of communication through which a machine can understand and extract information from hand sign images or video frames. It may act as a powerful interface between deaf-mute persons and the world around them by enabling communication with other human beings and computers without conventional input devices. However, development of such human-computer interaction (HCI) systems, specifically for deaf-mute communities, is a non-trivial research problem. Many studies and research works in this direction have been reported in literature. A few articles on survey or review of the reported methods are also available [1][2][3][4]. Hand gesture recognition, as evident in literature, may be broadly categorized into two groups, (i) sensor-based methods [5-9] and (ii) vision-based methods [10][11][12][13][14][15][16][17]. The sensor-based methods employ different kind of sensors to acquire hand posture images or videos. Depth sensor is the most popular among them. Using depth sensors such as Microsoft Kinect, Asus Xtion Pro, Intel Real Sense, depth information for object pixels at various locations is acquired, which is often useful in segmenting the region of interest (ROI), i.e., the hand portion containing the palm and the fingers from the acquired images containing surroundings as well. However, such sensors are not popular and economic, and hence, its usability is limited in comparison to vision-based systems which work with RGB images that are pervasively used at population scale.Vision-based gesture recognition has received significant attention of researchers for wide availability and usage of digital camera since the last two decades. Even cellphones, nowadays, have built-in high-resolution digital cameras which acquire RGB images. Moreover, web-cam integrated with laptop or attached with desktop

show abstract

Duodepth: Static Gesture Recognition Via Dual Depth Sensors

Cited by 2 publications

References 23 publications

A PointNet-Based Solution for 3D Hand Gesture Recognition

A PointNet-Based Solution for 3D Hand Gesture Recognition

Hand Signs Recognition from Cellphone Camera Captured Images for Deaf-Mute Persons

Contact Info

Product

Resources

About