ABSTRACT
This research is a study of the evaluation of full-body sketches and the principle of the human pose estimation using the OpenPose library, a method to detect 18 keypoints on a human structure. The dataset used in this research was drawing sketches of 22 first-year students, each of whom drew three drawings of three models. Detected keypoints are calculated to determine the angle and distance between keypoints, which provides 26 features. These features were modeled using ANN for predicting the grades of drawings classified as good, moderate, poor. The resulting keypoints are then taken to find the angles and distances of the skeleton, extracting 26 features and taking these features to create a model using ANN classification. The performance of the model was evaluated using with 56% accuracy
- “How to Draw People ebook by Jeff Mellem,” Rakuten Kobo. https://www.kobo.com/gr/en/ebook/how-to-draw-people-4 (accessed Nov. 22, 2020).Google Scholar
- L. Sigal, “Human Pose Estimation,” in Computer Vision: A Reference Guide, K. Ikeuchi, Ed. Boston, MA: Springer US, 2014, pp. 362–370. doi: 10.1007/978-0-387-31439-6_584.Google ScholarCross Ref
- Z. Cao, T. Simon, S.-E. Wei, and Y. Sheikh, “Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields,” ArXiv161108050 Cs, Nov. 2016, Accessed: Mar. 28, 2021. [Online]. Available: http://arxiv.org/abs/1611.08050Google Scholar
- CMU-Perceptual-Computing-Lab/openpose. CMU-Perceptual-Computing-Lab, 2021. Accessed: Mar. 28, 2021. [Online]. Available: https://github.com/CMU-Perceptual-Computing-Lab/openposeGoogle Scholar
- B. BOULAY, “Human posture recognition for behaviour understanding,” 2007.Google Scholar
- G. Guo and A. Lai, “A survey on still image based human action recognition,” Pattern Recognit., vol. 47, no. 10, pp. 3343–3361, Oct. 2014, doi: 10.1016/j.patcog.2014.04.018.Google ScholarCross Ref
- B. R. N, A. Subramanian, K. Ravichandran, and N. Venkateswaran, “Exploring Techniques to Improve Activity Recognition using Human Pose Skeletons,” in 2020 IEEE Winter Applications of Computer Vision Workshops (WACVW), Mar. 2020, pp. 165–172. doi: 10.1109/WACVW50321.2020.9096918.Google ScholarCross Ref
- S. Ghazal and U. S. Khan, “Human posture classification using skeleton information,” in 2018 International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), Mar. 2018, pp. 1–4. doi: 10.1109/ICOMET.2018.8346407.Google ScholarCross Ref
- Y. Agrawal, Y. Shah, and A. Sharma, “Implementation of Machine Learning Technique for Identification of Yoga Poses,” in 2020 IEEE 9th International Conference on Communication Systems and Network Technologies (CSNT), Apr. 2020, pp. 40–43. doi: 10.1109/CSNT48778.2020.9115758.Google ScholarCross Ref
- Z. Cao, G. Hidalgo, T. Simon, S.-E. Wei, and Y. Sheikh, “OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields,” ArXiv181208008 Cs, May 2019, Accessed: Nov. 26, 2020. [Online]. Available: http://arxiv.org/abs/1812.08008Google Scholar
- X. Chen, Z. Zhou, Y. Ying, and D. Qi, “Real-time Human Segmentation using Pose Skeleton Map,” in 2019 Chinese Control Conference (CCC), Jul. 2019, pp. 8472–8477. doi: 10.23919/ChiCC.2019.8865151.Google ScholarCross Ref
- “OpenCV,” OpenCV. https://opencv.org/ (accessed Mar. 28, 2021).Google Scholar
- V. Le, T. Nguyen, N. Tran, and T. Pham, “OpenPose's Evaluation in The Video Traditional Martial Arts Presentation,” in 2019 19th International Symposium on Communications and Information Technologies (ISCIT), Sep. 2019, pp. 76–81. doi: 10.1109/ISCIT.2019.8905243.Google ScholarCross Ref
- M. Hampton, Figure Drawing: Design and Invention, 2nd edition. United States? Michael Hampton, 2009.Google Scholar
- Q. Dang, J. Yin, B. Wang, and W. Zheng, “Deep learning based 2D human pose estimation: A survey,” Tsinghua Sci. Technol., vol. 24, no. 6, pp. 663–676, Dec. 2019, doi: 10.26599/TST.2018.9010100.Google ScholarCross Ref
- G. Ning, P. Liu, X. Fan, and C. Zhang, “A Top-down Approach to Articulated Human Pose Estimation and Tracking,” ArXiv190107680 Cs, Jan. 2019, Accessed: Nov. 24, 2020. [Online]. Available: http://arxiv.org/abs/1901.07680Google Scholar
- A. G. Howard , “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,” ArXiv170404861 Cs, Apr. 2017, Accessed: Feb. 23, 2021. [Online]. Available: http://arxiv.org/abs/1704.04861Google Scholar
- R. Bala and D. D. Kumar, “Classification Using ANN: A Review,” 2017. /paper/Classification-Using-ANN-%3A-A-Review-Bala-Kumar/cb6e2ae2427f183d6d49b2be5c3fd3b5bfef61c9 (accessed Mar. 29, 2021).Google Scholar
- TensorFlow, “Real-time Human Pose Estimation in the Browser with TensorFlow.js,” Medium, Sep. 27, 2018. https://medium.com/tensorflow/real-time-human-pose-estimation-in-the-browser-with-tensorflow-js-7dd0bc881cd5 (accessed Feb. 09, 2021).Google Scholar
- Paulo, “ProgrammingAI: Calculating the bearing between two points,” ProgrammingAI, Mar. 13, 2012. http://programmingai.blogspot.com/2012/03/calculating-bearing-between-two-points.html (accessed Jan. 26, 2021).Google Scholar
- R. Paul, “Euclidean Distance and Normalization of a Vector,” Medium, Dec. 23, 2020. https://medium.com/nerd-for-tech/euclidean-distance-and-normalization-of-a-vector-76f7a97abd9 (accessed Feb. 03, 2021).Google Scholar
Index Terms
- An Application of Evaluation of Human Sketches using Deep Learning Technique
Recommendations
Summarizing data using bottom-k sketches
PODC '07: Proceedings of the twenty-sixth annual ACM symposium on Principles of distributed computingA Bottom-sketch is a summary of a set of items with nonnegative weights that supports approximate query processing. A sketch is obtained by associating with each item in a ground set an independent random rank drawn from a probability distribution that ...
Human Pose Estimation using Deep Learning Techniques
ICIMMI '22: Proceedings of the 4th International Conference on Information Management & Machine IntelligenceHuman Pose Estimation is a method of detecting and classifying all the key human body joints. A 2D Human Pose Estimation method predicts the two-dimensional coordinates from an input image. Classical methods used deformable parts models for detecting ...
Comments