Exploring Techniques to Improve Activity Recognition using Human Pose Skeletons

Raj, Nawin; Subramanian, Ashwanth; Ravichandran, K.; Venkateswaran, N.

doi:10.1109/wacvw50321.2020.9096918

Cited by 4 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 7 illustrates the resulted average accuracies. [19] 91.1 98.3 −1.0 KcWKNN (on WVU) [19] 92.4 98.7 −1.1 3D Convolutional [21] 82.30 90.40 9.84 Convolutional Two Stream Fusion [23] 92.50 94.20 1.84 Improved two streams architecture [25] 94.00 94.20 0.21 Two Stream Fusion Convolutional [26] 92.70 93.60 0.97 Convolutional (ResNet-101) [24] 67.96 84.44 24.25 Two Stream Fusion MLP-LSTM [30] 79.21 96.92 22.36 Convnet conv-Lstm [27] 75.40 77.90 3.32 As shown in Table 7, our proposed technique achieved an acceptable level of performance in terms of average accuracy in almost every state-of-the-art classifier in the domain of video-based activity recognition. Indeed, the growth method achieved such performance with a significantly smaller number of required features to identify activities.…”

Section: Performance Indicatorsmentioning

confidence: 99%

A Novel Feature-Selection Method for Human Activity Recognition in Videos

et al. 2022

View full text Add to dashboard Cite

Human Activity Recognition (HAR) is the process of identifying human actions in a specific environment. Recognizing human activities from video streams is a challenging task due to problems such as background noise, partial occlusion, changes in scale, orientation, lighting, and the unstable capturing process. Such multi-dimensional and none-linear process increases the complexity, making traditional solutions inefficient in terms of several performance indicators such as accuracy, time, and memory. This paper proposes a technique to select a set of representative features that can accurately recognize human activities from video streams, while minimizing the recognition time and memory. The extracted features are projected on a canvas, which keeps the synchronization property of the spatiotemporal information. The proposed technique is developed to select the features that refer only to progression of changes. The original RGB frames are preprocessed using background subtraction to extract the subject. Then the activity pattern is extracted through the proposed Growth method. Three experiments were conducted; the first experiment was a baseline to compare the classification task using the original RGB features. The second experiment relied on classifying activities using the proposed feature-selection method. Finally, the third experiment provided a sensitivity analysis that compares between the effect of both techniques on time and memory resources. The results indicated that the proposed method outperformed original RBG feature-selection method in terms of accuracy, time, and memory requirements.

show abstract

Section: Performance Indicatorsmentioning

confidence: 99%

A Novel Feature-Selection Method for Human Activity Recognition in Videos

et al. 2022

View full text Add to dashboard Cite

show abstract

“…To obtain feature vectors invariant to a rigid body and affine transformations and to increase the generalizability of our approach, we based our calculation on image output and not on raw keypoints coordinates (Fig. 1) [41].…”

Section: Clinical Measuresmentioning

confidence: 99%

Using 2D Video-based Pose Estimation for Automated Prediction of Autism Spectrum Disorders in Preschoolers

Kojovic

Natraj

Mohanty³

et al. 2021

Preprint

View full text Add to dashboard Cite

Clinical research in autism has recently witnessed promising digital phenotyping results, mainly focused on single feature extraction, such as gaze, head turn on name-calling or visual tracking of the moving object. The main drawback of these studies is the focus on relatively isolated behaviors elicited by largely controlled prompts. We recognize that while the diagnosis process understands the indexing of the specific behaviors, ASD also comes with broad impairments that often transcend single behavioral acts. For instance, the atypical nonverbal behaviors manifest through global patterns of atypical postures and movements, fewer gestures used and often decoupled from visual contact, facial affect, speech. Here, we tested the hypothesis that a deep neural network trained on the non-verbal aspects of social interaction can effectively differentiate between children with ASD and their typically developing peers. Our model achieves an accuracy of 80.9% (F1 score: 0.818; precision: 0.784; recall: 0.854) with the prediction probability positively correlated to the overall level of symptoms of autism in social affect and repetitive and restricted behaviors domain. Provided the non-invasive and affordable nature of computer vision, our approach carries reasonable promises that a reliable machine-learning-based ASD screening may become a reality not too far in the future.

show abstract