Action Recognition Using Discriminative Structured Trajectory Groups

Atmosukarto, Indriyati; Ahuja, Narendra; Ghanem, Bernard

doi:10.1109/wacv.2015.124

Cited by 13 publications

(5 citation statements)

References 24 publications

(52 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In all our experiments, we first learn a GMM with 512 components and reduce the dimensionality of the final encoding to half using PCA. This is similar to the procedures in [39,3,4].…”

Section: Video Representationsupporting

confidence: 57%

ActivityNet: A large-scale video benchmark for human activity understanding

Heilbron

Escorcia

Ghanem

et al. 2015

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

1,727

816

View full text Add to dashboard Cite

“…In all our experiments, we first learn a GMM with 512 components and reduce the dimensionality of the final encoding to half using PCA. This is similar to the procedures in [39,3,4].…”

Section: Video Representationsupporting

confidence: 57%

ActivityNet: A large-scale video benchmark for human activity understanding

Heilbron

Escorcia

Ghanem

et al. 2015

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

1,727

816

View full text Add to dashboard Cite

“…The hierarchy was first computed using a divisive clustering algorithm and then represented a BOF-liked structure where each node was modeled by a bag-of-feature over MBH descriptors. In [28], the video was structured using bag of trajectory groups with latent class variables and employing multiple instance learning (MIL)-based support vector machine (SVM) to classification. More recently, Vig et al [29] revealed that the HAR performance can be maintained with as little as 30-40% of descriptors picked at random within dense trajectory-based approach [23].…”

Section: Related Workmentioning

confidence: 99%

Action recognition using length-variable edge trajectory and spatio-temporal motion skeleton descriptor

Weng

Guan

2018

J Image Video Proc.

View full text Add to dashboard Cite

Representing the features of different types of human action in unconstrained videos is a challenging task due to camera motion, cluttered background, and occlusions. This paper aims to obtain effective and compact action representation with length-variable edge trajectory (LV-ET) and spatio-temporal motion skeleton (STMS). First, in order to better describe the long-term motion information for action representation, a novel edge-based trajectory extracting strategy is introduced by tracking edge points from motion without limiting the length of trajectory; the end of the tracking is depending not only on the optical flow field but also on the optical flow vector position in the next frame. So, we only make use of a compact subset of action-related edge points in one frame to generate length-variable edge trajectories. Second, we observe that different types of action have their specific trajectory. A new trajectory descriptor named spatio-temporal motion skeleton is introduced; first, the LV-ET is encoded using both orientation and magnitude features and then the STMS is computed by motion clustering. Comparative experimental results with three unconstrained human action datasets demonstrate the effectiveness of our method.

show abstract

“…Hence, there exists a wealth of papers in this particular domain. The most recent literature on HAR from videos include trajectory-based descriptors [1]- [3], spatiotemporal feature representations [4]- [6], feature encoding [7]- [9], and deep learning [10]- [12]. Reviewing the extensive list of video-based HAR studies, however, goes beyond the scope of this study and we refer the reader to [13], [14] for a collection of more comprehensive surveys on the topic.…”

Section: Related Workmentioning

confidence: 99%

“…The actigraph accelerometer data contains post-filtered "counts" for each of the axes. These counts quantify the frequency and intensity of the user's activity 3 . Using Troiano's cut point scale [64], the age of a user, and their accelerometer triaxial count, each epoch is labeled with an intensity level: Sedentary, Light, Moderate, or Vigorous.…”

Section: Activity Mode Detectionmentioning

confidence: 99%

Robust Automated Human Activity Recognition and Its Application to Sleep Research

Sathyanarayana

Ofli

Fernández-Luque

et al. 2016

2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW)

View full text Add to dashboard Cite

Abstract-Human Activity Recognition (HAR) is a powerful tool for understanding human behaviour. Applying HAR to wearable sensors can (1) provide new insights by enriching the feature set in health studies, and (2) enhance the personalisation and effectiveness of health, wellness, and fitness applications. Wearable devices provide an unobtrusive platform for user monitoring, and due to their increasing market penetration, feel intrinsic to the wearer. The integration of these devices in daily life provide a unique opportunity for understanding human health and wellbeing. This is referred to as the "quantified self" movement.The analyses of complex health behaviours such as sleep, traditionally require a time-consuming manual interpretation by experts. This manual work is necessary due to the erratic periodicity and persistent noisiness of human behaviour. In this paper, we present a robust automated human activity recognition algorithm, which we call RAHAR. We test our algorithm in the application area of sleep research by providing a novel framework for evaluating sleep quality and examining the correlation between the aforementioned and an individual's physical activity. Our results improve the state-of-the-art procedure in sleep research by 15% for area under ROC and by 30% for F1 score on average. However, application of RAHAR is not limited to sleep analysis and can be used for understanding other health problems such as obesity, diabetes, and cardiac diseases.

show abstract

Action Recognition Using Discriminative Structured Trajectory Groups

Cited by 13 publications

References 24 publications

ActivityNet: A large-scale video benchmark for human activity understanding

ActivityNet: A large-scale video benchmark for human activity understanding

Action recognition using length-variable edge trajectory and spatio-temporal motion skeleton descriptor

Robust Automated Human Activity Recognition and Its Application to Sleep Research

Contact Info

Product

Resources

About