Segmentation and Recognition of Continuous Human Activity

Ali, Asad; Aggarwal, J. K.

doi:10.21236/ada396147

Cited by 16 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many well-known skeletonization techniques, such as thinning, Voronoi diagrams, and the distance transform. However, these approaches are computationally expensive and highly sensitive to noise in the target silhouette [9]. Therefore, approaches those are inexpensive and relatively robust against noise have been developed to detect the extreme points of a target's silhouette in order to produce a "star" skeleton [6,15].…”

Section: Convex Polygon-based Star Figure Representationmentioning

confidence: 99%

“…Moment based features were extracted from MEI and MHI and were used to conduct template matching. It is also possible to perform human behavior analysis based on human silhouette analysis [6][7][8][9]. In [7], the projection histograms of each person were computed and compared with the probabilistic projection maps stored for each posture during the training phase.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Atomic Human Action Segmentation and Recognition Using a Spatio-Temporal Probabilistic Framework

Chen

Liao

Shih

2007

Int. J. Semantic Computing

View full text Add to dashboard Cite

In this paper, a framework of automatic human action segmentation and recognition in continuous action sequences is proposed. A star figure enclosed by a bounding convex polygon is used to effectively represent the extremities of the silhouette of a human body. The human action, thus, is recorded as a sequence of the star-figure's parameters, which is used for action modeling. To model human actions in a compact manner while characterizing their spatio-temporal distributions, the star-figure's parameters are represented by Gaussian mixture models (GMM). In addition, to address the intrinsic nature of temporal variations in a continuous action sequence, we transform the time sequence of star-like figure parameters into frequency domain by discrete cosine transform (DCT) and use only the first few coefficients to represent different temporal patterns with significant discriminating power. The performance shows that the proposed framework can recognize continuous human actions in an efficient way.

show abstract

Section: Convex Polygon-based Star Figure Representationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Atomic Human Action Segmentation and Recognition Using a Spatio-Temporal Probabilistic Framework

Chen

Liao

Shih

2007

Int. J. Semantic Computing

View full text Add to dashboard Cite

show abstract

“…Similar work has been pursued in activity segmentation where a continuous stream of sensor data is segmented into discrete meaningful units that can be separately classified. Approaches have been tried that train supervised learning algorithms based on examples of activity starts and stops, examples of activity transitions, or examples of activity pairs [40], [41]. Lack of confidence in the activity classification has also been used to indicate a possible transition between activities [42].…”

Section: Introductionmentioning

confidence: 99%

Automated Detection of Activity Transitions for Prompting

Feuz

Cook

Rosasco

et al. 2015

IEEE Trans. Human-Mach. Syst.

View full text Add to dashboard Cite

Individuals with cognitive impairment can benefit from intervention strategies like recording important information in a memory notebook. However, training individuals to use the notebook on a regular basis requires a constant delivery of reminders. In this work, we design and evaluate machine learning-based methods for providing automated reminders using a digital memory notebook interface. Specifically, we identify transition periods between activities as times to issue prompts. We consider the problem of detecting activity transitions using supervised and unsupervised machine learning techniques, and find that both techniques show promising results for detecting transition periods. We test the techniques in a scripted setting with 15 individuals. Motion sensors data is recorded and annotated as participants perform a fixed set of activities. We also test the techniques in an unscripted setting with 8 individuals. Motion sensor data is recorded as participants go about their normal daily routine. In both the scripted and unscripted settings a true positive rate of greater than 80% can be achieved while maintaining a false positive rate of less than 15%. On average, this leads to transitions being detected within 1 minute of a true transition for the scripted data and within 2 minutes of a true transition on the unscripted data.

show abstract

“…Vision based gesture analysis has been studied for providing an alternative interaction between human and computer in recent years [1][2][3][4][5][6][7][8][9][10]. Especially, gesture recognition has been major research topic for a natural human-robot interaction.…”

Section: Introductionmentioning

confidence: 99%

Gesture analysis for human-robot interaction

Kim¹,

Kwak²,

Young³

2006

2006 8th International Conference Advanced Communication Technology

View full text Add to dashboard Cite

    This paper is to present gesture analysis for human-robot interaction. Gesture analysis is consisted of four processes such as detecting of hand in bimanual movements, splitting of a meaning gesture region from image stream, extracting features and recognizing gesture. Skin color analysis, image motion detection and shape information are used to detect bimanual hand movements and gesture spotting. Skin color information for tracking hand gesture is obtained from face detection region. The velocity of moving hand is calculated for detecting a meaning gesture region from consecutive image frames. Combined gesture features such as structural and statistical features are extracted from image stream. We have experimented to evaluate detection of bimanual hand movements and gesture recognition with a camera, which is pan/tilt and a single camera that is mounted on mobile robot. Performance evaluation of gesture recognition has experimented using ETRI database and an encouraging recognition rate of 89% has been obtained.

show abstract

Segmentation and Recognition of Continuous Human Activity

Cited by 16 publications

References 0 publications

Atomic Human Action Segmentation and Recognition Using a Spatio-Temporal Probabilistic Framework

Atomic Human Action Segmentation and Recognition Using a Spatio-Temporal Probabilistic Framework

Automated Detection of Activity Transitions for Prompting

Gesture analysis for human-robot interaction

Contact Info

Product

Resources

About