View-independent Hand Posture Recognition from Single Depth Images Using PCA and Flusser Moments

Gallo, Luigi; Placitelli, Alessio Pierluigi

doi:10.1109/sitis.2012.133

Cited by 8 publications

(9 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper significantly extends the approach proposed in [4], in which a method for extracting and processing region-based statistical image features and a database containing pose compensated depth images of hands were presented. In more detail, the main contributions of this work are: i) a novel weighting method, which takes advantage of the velocity and orientation of the user's hand with respect to the sensor to improve the hand shape classification accuracy; and, ii) a user study, the results of which show that the velocity-based weighting approach increases the accuracy of the hand shape classification while performing 3D interaction tasks.…”

Section: Introductionmentioning

confidence: 78%

Hand shape classification using depth data for unconstrained 3D interaction

Gallo

2014

Journal of Ambient Intelligence and Smart Environments

View full text Add to dashboard Cite

In this paper, we introduce a novel method for view-independent hand pose recognition from depth data. The proposed approach, which does not rely on color information, provides an estimation of the shape and orientation of the user's hand without constraining him/her to maintain a fixed position in the 3D space. We use principal component analysis to estimate the hand orientation in space, Flusser moment invariants as image features and two SVM-RBF classifiers for visual recognition. Moreover, we describe a novel weighting method that takes advantage of the orientation and velocity of the user's hand to assign a score to each hand shape hypothesis. The complete processing chain is described and evaluated in terms of real-time performance and classification accuracy. As a case study, it has also been integrated into a touchless interface for 3D medical visualization, which allows users to manipulate 3D anatomical parts with up to six degrees of freedom. Furthermore, the paper discusses the results of a user study aimed at assessing if using hand velocity as an indicator of the user's intentionality in changing hand posture results in an overall gain in the classification accuracy. The experimental results show that, especially in the presence of out-of-plane rotations of the hand, the introduction of the velocity-based weighting method produces a significant increase in the pose recognition accuracy.

show abstract

Section: Introductionmentioning

confidence: 78%

Hand shape classification using depth data for unconstrained 3D interaction

Gallo

2014

Journal of Ambient Intelligence and Smart Environments

View full text Add to dashboard Cite

show abstract

“…The ESF descriptor consists of concatenated histograms which are generated with the random points in the point cloud. In addition to the features in the original depth domain, Haarlet coefficients [45], Gabor coefficients [88] and Flusser moment invariants [113,114] have been extracted from transformed hand images to efficiently form rotation and intensity invariant hand features. In contrast to low-level hand features which are based on the global hand information, the mid-level hand features are usually based on the local patch-level descriptors [115].…”

Section: A Feature Representationmentioning

confidence: 99%

Survey on 3D Hand Gesture Recognition

Cheng

Yang

Liu

2016

IEEE Trans. Circuits Syst. Video Technol.

290

110

View full text Add to dashboard Cite

3D hand gesture recognition has attracted increasing research interests in computer vision, pattern recognition and human-computer interaction. The emerging depth sensors greatly inspired various hand gesture recognition approaches and applications, which were severely limited in the 2D domain with conventional cameras. This paper presents a survey of some recent works on hand gesture recognition using 3D depth sensors. We first review the commercial depth sensors and public datasets which are widely used in this field. Then, we review the state-of-the-art research for 3D hand gesture recognition in four aspects: 3D hand modeling, static hand gesture recognition, hand trajectory gesture recognition and continuous hand gesture recognition. While the emphasis is on 3D hand gesture recognition approaches, the related applications and typical systems are also briefly summarized for practitioners.

show abstract

“…Specifically, the Kinect 6+3 DOF interface allows the user to control separately the rotation/translation of the box clipping and the rotation of the volume. The system is able to discriminate between the open palm and clenched fist of the dominant and non-dominant hand [Gallo and Placitelli 2012], and to track their positions with respect to the trunk. When the non-dominant arm is outstretched more than 55% of its total length and the palm is open, the system synchronizes the box cropping with the user's hand: by changing the position or the orientation of the dominant hand, the position and the orientation of the box cropping change accordingly.…”

Section: Interfacesmentioning

confidence: 99%

A study on the degrees of freedom in touchless interaction

Gallo

2013

SIGGRAPH Asia 2013 Technical Briefs

View full text Add to dashboard Cite

During the last few years, we have been witnessing a widespread adoption of touchless technologies in the context of surgical procedures. Touchless interfaces are advantageous in that they can preserve sterility around the patient, allowing surgeons to visualize medical images without having to physically touch any control or to rely on a proxy. Such interfaces have been tailored to interact with 2D medical images but not with 3D reconstructions of anatomical data, since such an interaction requires at least three degrees of freedom. In this paper, we discuss the results of a user study in which a mouse-based interface has been compared with two Kinect-based touchless interfaces which allow users to interact with 3D data with up to nine degrees of freedom. The experimental results show that there is a significant relation between the number of degrees of freedom simultaneously controlled by the user and the number of degrees of freedom required to perform, in a touchless way, an accurate manipulation task.

show abstract

View-independent Hand Posture Recognition from Single Depth Images Using PCA and Flusser Moments

Cited by 8 publications

References 22 publications

Hand shape classification using depth data for unconstrained 3D interaction

Hand shape classification using depth data for unconstrained 3D interaction

Survey on 3D Hand Gesture Recognition

A study on the degrees of freedom in touchless interaction

Contact Info

Product

Resources

About