Machine Learning for Real Time Poses Classification Using Kinect Skeleton Data

Choubik, Youness; Mahmoudi, Abdelhak

doi:10.1109/cgiv.2016.66

Cited by 18 publications

(7 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many researchers who apply machine learning algorithms on Kinect data for pose and gesture classification, e.g. [48,49]. Since Kinect V1 cannot detect finer movements and motion capture is still not 100% accurate, applying learning algorithms can ameliorate the results.…”

Section: Methodsmentioning

confidence: 99%

An Approach to Develop a Motion-Sensitive, Locally Multiplayer-Hybrid (Multipid) 3d Video Game

Kayar

2019

Mugla Journal of Science and Technology

View full text Add to dashboard Cite

Playing video games has been gaining increased interest over the last two decades regardless of age, gender and any kind of status. Video games are generally controlled via frequently used input devices such as game controllers or mouse and keyboard. However, there are also newer technologies which can detect human motion through sensors. This technology enhances the game-play experience due to its interactivity level and therefore, such an experience embraces many players more than traditional controllers. The term "Hybrid" is used for both the multi-genre games and for games integrating conventional controllers together with new technology devices, e.g. motion-sensitive devices. In this paper, we use the term for the latter. Our aim is to provide the game developers a roadmap to generate a 3D hybrid game in order to enhance the game-play experience where we mainly control the game with keyboard but also integrate Kinect to sense human motion and solve interactive puzzles at the end of each level. Besides, we also suggest to enable the multiplayer mode for some interactive puzzles to challenge the players and observe their contentedness in terms of game-play. After this experience, we named this game model as MultipId game-model as it combines the features of both Multiplayer and Hybrid modes. We developed a game employing this model and conducted a user study which is discussed in the Results section. When we asked the players, more than 60% have stated that Kinect-based local multiplayer subsections improve the engagement of game quite effectively. Besides, 87% of the attendees stated that they like to see more MultipId games in the market. We believe that increasing the number of MultipId games gives a competitive edge over existing single-player games that employ only traditional controllers.

show abstract

Section: Methodsmentioning

confidence: 99%

An Approach to Develop a Motion-Sensitive, Locally Multiplayer-Hybrid (Multipid) 3d Video Game

Kayar

2019

Mugla Journal of Science and Technology

View full text Add to dashboard Cite

show abstract

“…This solution gives complete body information, including articulations of the upper and lower limbs, without needing a depth sensor camera or fiducial markers on the body. Choubik and Mahmoudi (2016) have successfully classified human poses using a feature vector calculated from the Kinect skeleton structure. The vocabulary of the classifier had 18 poses associated with both arms.…”

Section: Motion Capture Approachesmentioning

confidence: 99%

Section: Skeleton Data and Preprocessingmentioning

confidence: 99%

“…The rotation format keeps relative value between the points, similar to the difference between points seen in related work (Choubik andMahmoudi, 2016; Ijjina andMohan, 2014). Choubik and Mahmoudi (2016) use the difference between the joints to classify the person's pose. For example, values calculated for variable d that means the distance between shoulder and wrist would result in d = 100px if the person is on the left or on the right side of the frame (Figure 4), which is correct since the position relative to the camera does not influence which pose the person is doing.…”

Section: Skeleton Data and Preprocessingmentioning

confidence: 99%

See 1 more Smart Citation

Supervised Classification of Motor-Rehabilitation Body Movements with RGB Cameras and Pose Tracking Data

Rodrigues

Dias²,

Guimarães³

et al. 2022

JIS

View full text Add to dashboard Cite

The technological evolution allowed the use of a single camera for precise and effective body tracking, reducing the cost and increasing the accessibility of applications in places where depth cameras and wearable sensors are not available. This paper describes and implements a supervised machine learning process consisting of a mobile application used as a motion capture device which also transforms the data into an input for a machine learning model that classifies upper and lower limbs movements (24 types of human movements). The user performs movements in front of the camera, and the trained model classifies them. We designed the system to work in a motor-rehabilitation context to assist the professional while the patient does physical exercises. The implementation can summarize the movements made during the rehabilitation sessions by counting the repetitions and classifying them when done completely or reached a specific range of motion.

show abstract

“…In visionbased action recognition, the common approach is to extract image features from video data and to issue a corresponding action class label (Poppe, 2010;Babiker et al, 2018). Nevertheless, when skeleton representation of the human body is used, the most privileged discriminative features are the raw data coming from the skeletal tracking (joint spatial coordinates) (Patsadu et al, 2012;Youness and Abdelhak, 2016) or some indices expressing geometric relations between certain body points, such as: the vertical distance from hip joint to room floor (Visutarrom et al, 2014(Visutarrom et al, , 2015, the distance between the right toe and the plane spanned by the left ankle, the left hip and the foot for a fixed pose (Müller et al, 2005) the distance between two joints, two body segments, or a joint and a body segment (Yang and Tian, 2014), the relative angle between two segments within the body kinematic chain (Müller et al, 2005) and finally, the size of the 3D bounding box enclosing the body skeleton (Bevilacqua et al, 2014). Geometric features are synthetic in the sense that they express a single geometric aspect making them particularly robust to spatial variations that are not correlated with the aspect of interest (Müller et al, 2005).…”

Section: Introductionmentioning

confidence: 99%

Automatic Pose Recognition for Monitoring Dangerous Situations in Ambient-Assisted Living

Guerra

Ramat

Beltrami

et al. 2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Continuous monitoring of frail individuals for detecting dangerous situations during their daily living at home can be a powerful tool toward their inclusion in the society by allowing living independently while safely. To this goal we developed a pose recognition system tailored to disabled students living in college dorms and based on skeleton tracking through four Kinect One devices independently recording the inhabitant with different viewpoints, while preserving the individual's privacy. The system is intended to classify each data frame and provide the classification result to a further decisionmaking algorithm, which may trigger an alarm based on the classified pose and the location of the subject with respect to the furniture in the room. An extensive dataset was recorded on 12 individuals moving in a mockup room and undertaking four poses to be recognized: standing, sitting, lying down, and "dangerous sitting." The latter consists of the subject slumped in a chair with his/her head lying forward or backward as if unconscious. Each skeleton frame was labeled and represented using 10 discriminative features: three skeletal joint vertical coordinates and seven relative and absolute angles describing articular joint positions and body segment orientation. In order to classify the pose of the subject in each skeleton frame we built a two hidden layers multi-layer perceptron neural network with a "SoftMax" output layer, which we trained on the data from 10 of the 12 subjects (495,728 frames), with the data from the two remaining subjects representing the test set (106,802 frames). The system achieved very promising results, with an average accuracy of 83.9% (ranging 82.7 and 94.3% in each of the four classes). Our work proves the usefulness of human pose recognition based on machine learning in the field of safety monitoring in assisted living conditions.

show abstract

Machine Learning for Real Time Poses Classification Using Kinect Skeleton Data

Cited by 18 publications

References 1 publication

An Approach to Develop a Motion-Sensitive, Locally Multiplayer-Hybrid (Multipid) 3d Video Game

An Approach to Develop a Motion-Sensitive, Locally Multiplayer-Hybrid (Multipid) 3d Video Game

Supervised Classification of Motor-Rehabilitation Body Movements with RGB Cameras and Pose Tracking Data

Automatic Pose Recognition for Monitoring Dangerous Situations in Ambient-Assisted Living

Contact Info

Product

Resources

About