YogaTube: A Video Benchmark for Yoga Action Recognition

Yadav, Santosh Kumar; Singh, Guntaas; Verma, Manisha; Tiwari, Kamlesh; Pandey, Hari Mohan; Akbar, Shaik; Corcoran, Peter

doi:10.1109/ijcnn55064.2022.9892122

Cited by 7 publications

(6 citation statements)

References 43 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Yoga asana image dataset is important for improving Yoga practice, enhancing teaching, supporting research, advancing technology, and promoting accessibility. Previous research works have created Yoga datasets, however those datasets do not cover effective and ineffective ways of doing yoga asana [ 13 , 14 ]. Moreover, these datasets represents various yoga poses but not the complete Yoga asana steps from start to end.…”

Section: Data Descriptionmentioning

confidence: 99%

Yoga dataset: A resource for computer vision-based analysis of Yoga asanas

Suryawanshi,

Gunjal,

Kanorewala

et al. 2023

Data in Brief

View full text Add to dashboard Cite

Section: Data Descriptionmentioning

confidence: 99%

Yoga dataset: A resource for computer vision-based analysis of Yoga asanas

Suryawanshi,

Gunjal,

Kanorewala

et al. 2023

Data in Brief

View full text Add to dashboard Cite

“…YogaTube [3] introduces a new large-scale video benchmark dataset for yoga action recognition, featuring 82 classes and 5,484 samples. It includes RGB videos, optical flow data, and skeleton data from sources such as YouTube, Bing, Flickr, and more.…”

Section: Yoga Datasetsmentioning

confidence: 99%

“…In recent times, the emergence of professional sports datasets [1][2][3][5][6][7][8][9][10][11][12][13][14] has played a vital role in advancing research and applications in the field of health and wellbeing. Notably, the task of recognizing yoga poses presents a unique challenge, as it involves a diverse range of continuous movements, even within a single posture.…”

Section: Introductionmentioning

confidence: 99%

A Study on the Risk Assessment Tool of Terrorism in Jung-gu Gwanghwamun Area, Seoul, South Korea

Kim¹,

Cho²,

Park³

2021

Korean Security Science Review

View full text Add to dashboard Cite

The increasing popularity of exercises including yoga and Pilates has created a greater demand for professional exercise video datasets in the realm of artificial intelligence. In this study, we developed 3DYoga90 1 , which is organized within a three-level label hierarchy. We have expanded the number of poses from an existing state-ofthe-art dataset, increasing it from 82 to 90 poses. Our dataset includes meticulously curated RGB yoga pose videos and 3D skeleton sequences. This dataset was created by a dedicated team of six individuals, including yoga instructors. It stands out as one of the most comprehensive open datasets, featuring the largest collection of RGB videos and 3D skeleton sequences among publicly available resources. This contribution has the potential to significantly advance the field of yoga action recognition and pose assessment. Additionally, we conducted experiments to evaluate the practicality of our proposed dataset. We employed three different model variants for benchmarking purposes.

show abstract

“…The authors compare Resnet, DenseNet, MobileNet, and ResNext models on their proposed dataset and conclude that DenseNet-201 outperforms other models. Moreover, Yadav et al [17] creates a video dataset including 5484 videos in 82 classes of yoga poses. They also propose an architecture with three parallel components which use the part affinity fields model, optical flow, and raw RGB videos to classify yoga asanas.…”

Section: Introductionmentioning

confidence: 99%

Comparison of CNN-based methods for yoga pose classification

ATALAY AYDIN

2024

Turkish Journal of Engineering

View full text Add to dashboard Cite

Yoga is an exercise developed in ancient India. People perform yoga in order to have mental, physical, and spiritual benefits. While yoga helps build strength in the mind and body, incorrect postures might result in serious injuries. Therefore, yoga exercisers need either an expert or a platform to receive feedback on their performance. Since access to experts is not an option for everyone, a system to provide feedback on the yoga poses is required. To this end, commercial products such as smart yoga mats and smart pants are produced; Kinect cameras, sensors, and wearable devices are used. However, these solutions are either uncomfortable to wear or not affordable for everyone. Nonetheless, a system that employs computer vision techniques is a requirement. In this paper, we propose a deep-learning model for yoga pose classification, which is the first step of a quality assessment system. We introduce a wavelet-based model that first takes wavelet transform of input images. The acquired subbands, i.e., approximation, horizontal, vertical, and diagonal coefficients of the wavelet transform are then fed into separate convolutional neural networks (CNN). The obtained probability results for each group are fused in order to have the final yoga class prediction. A publicly available dataset with 5 yoga poses is used. Since the number of images in the dataset is not enough for a deep learning model, we also perform data augmentation to increase the number of images. We compare our results to a CNN model and the three models that employ the subbands separately. Results obtained using the proposed model outperforms the accuracy output achieved with the compared models.

show abstract

YogaTube: A Video Benchmark for Yoga Action Recognition

Cited by 7 publications

References 43 publications

Yoga dataset: A resource for computer vision-based analysis of Yoga asanas

Yoga dataset: A resource for computer vision-based analysis of Yoga asanas

A Study on the Risk Assessment Tool of Terrorism in Jung-gu Gwanghwamun Area, Seoul, South Korea

Comparison of CNN-based methods for yoga pose classification

Contact Info

Product

Resources

About