Focus-of-attention for human activity recognition from UAVs

Burghouts, Gertjan J.; Eekeren, Adam W. M. van; Dijk, J. C. van

doi:10.1117/12.2067569

Cited by 5 publications

(7 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The classification results are comparable with the ones showed in this research, and its better accuracy is the tradeoff with the simplistic implementation that is not capable of recognizing more objects or actions. The results presented in [114] have little more classes to be separated. The tracking activity is applied directly using BOW, which makes the architecture more rigid when adding new activities.…”

Section: Surveillance Missionmentioning

confidence: 88%

ARCog: An Aerial Robotics Cognitive Architecture

et al. 2020

View full text Add to dashboard Cite

SUMMARY Efficient algorithm integration is a key issue in aerial robotics. However, only a few integration solutions rely on a cognitive approach. Cognitive approaches break down complex problems into independent units that may deal with progressively lower-level data interfaces, all the way down to sensors and actuators. A cognitive architecture defines information flow among units to produce emergent intelligent behavior. Despite the improvements in autonomous decision-making, several key issues remain open. One of these issues is the selection, coordination, and decision-making related to the several specialized tasks required for fulfilling mission objectives. This work addresses decision-making for the cognitive unmanned-aerial-vehicle architecture coined as ARCog. The proposed architecture lays the groundwork for the development of a software platform aligned with the requirements of the state-of-the-art technology in the field. The system is designed to provide high-level decision-making. Experiments prove that ARCog works correctly in its target scenario.

show abstract

Section: Surveillance Missionmentioning

confidence: 88%

ARCog: An Aerial Robotics Cognitive Architecture

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Three of these activities were static, i.e., performed in place (waving, dogging throwing), while two were dynamic (walking and running). These five activities were selected to have a fair comparison with other works which used the same five activities for human activity classification [77,84,85]. The performance of human detection models and human activity classification models were evaluated using these five activities.…”

Section: A Dataset Overviewmentioning

confidence: 99%

“…The comparison was done in terms of activity classification accuracy. Burghouts et al [84] proposed the first HAR method that extracted motion features from videos and utilized these features for activity classification. This method yielded an accuracy of 57%.…”

Section: ) Human Activity Classification Experimentsmentioning

confidence: 99%

“…Human detection and activity recognition (HDAR) using the highly challenging UCF-ARG aerial dataset has been done using various methods [8,77,83,84,85]. The combination of "The Fastest Pedestrian Detector in the West" (FPDW) [114] and moving object detection was utilized for human detection and tracking in UAV-based videos [83].…”

Section: Introductionmentioning

confidence: 99%

“…Here, visual words were utilized to represent motion features, which were described as a frequency count of the words. The SVM classifier served as the activity detector [84]. Lastly, an automated UAVbased DL algorithm consisted of video stabilization using the surf feature selection and Lucas-Kanade method, human area detection using faster R-CNN, and action recognition using a structure combining a three-dimensional CNN architecture and a residual network [85].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Comparison Between Various Human Detectors and CNN-Based Feature Extractors for Human Activity Recognition via Aerial Captured Video Sequences

et al. 2022

View full text Add to dashboard Cite

Human detection and activity recognition (HDAR) in videos plays an important role in various real-life applications. Recently, object detection methods such as "you only look once" (YOLO), faster region based convolutional neural network (R-CNN), and EfficientDet have been used to detect humans in videos for subsequent decision-making applications. This paper aims to address the problem of human detection in aerial captured video sequences using a moving camera attached to an aerial platform with dynamical events such as varied altitudes, illumination changes, camera jitter, and variations in viewpoints, object sizes and colors. Unlike traditional datasets that have frames captured by a static ground camera with medium or large regions of humans in these frames, the UCF-ARG aerial dataset is more challenging because it contains videos with large distances between the humans in the frames and the camera. The performance of human detection methods that have been described in the literature are often degraded when input video frames are distorted by noise, blur, illumination changes, and the like. To address these limitations, the object detection methods used in this study were trained on the COCO dataset and evaluated on the publicly available UCF-ARG dataset. The comparison between these detectors was done in terms of detection accuracy. The performance evaluation considers five human actions (digging, waving, throwing, walking, and running). Experimental results demonstrated that EfficientDetD7 was able to outperform other detectors with 92.9% average accuracy in detecting all activities and various conditions including blurring, addition of Gaussian noise, lightening, and darkening. Additionally, deep pre-trained convolutional neural networks (CNNs) such as ResNet and EfficientNet were used to transfer learning from the ImageNet dataset to the UCF-ARG dataset and to extract highly informative features from the detected and cropped human patches. The extracted spatial features were utilized by Long Short-Term Memory (LSTM) to consider temporal relations between features for human activity recognition (HAR). Experimental results found that the EfficientNetB7-LSTM was able to outperform existing HAR methods in terms of average accuracy (80%), average precision (83%), average recall (80%), average F1 score (80%), average false negative rate (FNR) (20%), average false positive rate (FPR) (4.8%), and average Area Under Curve (AUC) (94%). The outcome is a robust HAR system which combines EfficientDetD7, and EfficientNetB7 with LSTM for human detection and activity classification, respectively.

show abstract

Fully Autonomous UAV-Based Action Recognition System Using Aerial Imagery

Han

Razi

2020

Advances in Visual Computing

View full text Add to dashboard Cite

Focus-of-attention for human activity recognition from UAVs

Cited by 5 publications

References 26 publications

ARCog: An Aerial Robotics Cognitive Architecture

ARCog: An Aerial Robotics Cognitive Architecture

A Comparison Between Various Human Detectors and CNN-Based Feature Extractors for Human Activity Recognition via Aerial Captured Video Sequences

Fully Autonomous UAV-Based Action Recognition System Using Aerial Imagery

Contact Info

Product

Resources

About