Human action recognition from RGB-D data using complete local binary pattern

Arivazhagan, S.; Shebiah, R. Newlin; Harini, R.; Swetha, S.

doi:10.1016/j.cogsys.2019.05.002

Cited by 29 publications

(14 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, VM signals in our study provided sufficient information for identifying children's activity intensities while indoors with good accuracy. Future studies using computer vision to classify activity intensities in children may consider the use of human activity recognition algorithms [29][30][31] to specifically target sedentary behaviors, such as quiet sitting, from motionless non-sedentary activities, such as quiet standing or performing an isometric bodyweight resistance exercise (e.g., a sustained squat). Furthermore, the estimation of object mass from images is a nontrivial problem [32], which requires additional attention in the computer vision physical activity literature with respect to differential activity intensity classification during human-object interactions.…”

Section: Discussionmentioning

confidence: 99%

Quantifying Physical Activity in Young Children Using a Three-Dimensional Camera

McCullough

Rodríguez

Garber

2020

Sensors

View full text Add to dashboard Cite

The purpose of this study was to determine the feasibility and validity of using three-dimensional (3D) video data and computer vision to estimate physical activity intensities in young children. Families with children (2–5-years-old) were invited to participate in semi-structured 20-minute play sessions that included a range of indoor play activities. During the play session, children’s physical activity (PA) was recorded using a 3D camera. PA video data were analyzed via direct observation, and 3D PA video data were processed and converted into triaxial PA accelerations using computer vision. PA video data from children (n = 10) were analyzed using direct observation as the ground truth, and the Receiver Operating Characteristic Area Under the Curve (AUC) was calculated in order to determine the classification accuracy of a Classification and Regression Tree (CART) algorithm for estimating PA intensity from video data. A CART algorithm accurately estimated the proportion of time that children spent sedentary (AUC = 0.89) in light PA (AUC = 0.87) and moderate-vigorous PA (AUC = 0.92) during the play session, and there were no significant differences (p > 0.05) between the directly observed and CART-determined proportions of time spent in each activity intensity. A computer vision algorithm and 3D camera can be used to estimate the proportion of time that children spend in all activity intensities indoors.

show abstract

Section: Discussionmentioning

confidence: 99%

Quantifying Physical Activity in Young Children Using a Three-Dimensional Camera

McCullough

Rodríguez

Garber

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Being able to detect body skeleton, this commercial device includes IR and RGB sensors to achieve an accurate pose estimation and movement capture. In the HRI domain, Mazhar et al [37] and Arivazhagan et al [38] achieved robust hand gesture detection using the 3D skeleton extraction feature of a Kinect device.…”

Section: Gesture Recognition For Hrimentioning

confidence: 99%

Comparison of RGB-D and IMU-based gesture recognition for human-robot interaction in remanufacturing

Roda-Sánchez

Garrido-Hidalgo

García

et al. 2021

Int J Adv Manuf Technol

View full text Add to dashboard Cite

With product life-cycles getting shorter and limited availability of natural resources, the paradigm shift towards the circular economy is being impulsed. In this domain, the successful adoption of remanufacturing is key. However, its associated process efficiency is to date limited given high flexibility requirements for product disassembly. With the emergence of Industry 4.0, natural human-robot interaction is expected to provide numerous benefits in terms of (re)manufacturing efficiency and cost. In this regard, vision-based and wearable-based approaches are the most extended when it comes to establishing a gesture-based interaction interface. In this work, an experimental comparison of two different movement-estimation systems—(i) position data collected from Microsoft Kinect RGB-D cameras and (ii) acceleration data collected from inertial measurement units (IMUs)—is addressed. The results point to our IMU-based proposal, OperaBLE, having recognition accuracy rates up to 8.5 times higher than these of Microsoft Kinect, which proved to be dependent on the movement’s execution plane, subject’s posture, and focal distance.

show abstract

“…The training phase of the presented approach is fast, robust and it does not require careful parameter tuning. In Reference [13] both RGB and Depth Camera are used to extract motion features, generating a Salient Information Map. For each motion history image, a Complete Local Binary descriptor is computed, extracting sign, magnitude and center descriptors from the Salient Information Map.…”

Section: Related Workmentioning

confidence: 99%

Using Common Spatial Patterns to Select Relevant Pixels for Video Activity Recognition

et al. 2020

View full text Add to dashboard Cite

Video activity recognition, despite being an emerging task, has been the subject of important research due to the importance of its everyday applications. Video camera surveillance could benefit greatly from advances in this field. In the area of robotics, the tasks of autonomous navigation or social interaction could also take advantage of the knowledge extracted from live video recording. In this paper, a new approach for video action recognition is presented. The new technique consists of introducing a method, which is usually used in Brain Computer Interface (BCI) for electroencephalography (EEG) systems, and adapting it to this problem. After describing the technique, achieved results are shown and a comparison with another method is carried out to analyze the performance of our new approach.

show abstract

Human action recognition from RGB-D data using complete local binary pattern

Cited by 29 publications

References 45 publications

Quantifying Physical Activity in Young Children Using a Three-Dimensional Camera

Quantifying Physical Activity in Young Children Using a Three-Dimensional Camera

Comparison of RGB-D and IMU-based gesture recognition for human-robot interaction in remanufacturing

Using Common Spatial Patterns to Select Relevant Pixels for Video Activity Recognition

Contact Info

Product

Resources

About