Overview of 2D Human Pose Estimation

Ruijun Ma ； Jian Zhang ； Yuming Qi

《World Scientific Research Journal》 6卷1期 (2020/01) Pp. 196-202

https://doi.org/10.6911/WSRJ.202001_6(1).0028

摘要

2D human pose estimation is the basic content of human pose estimation, mainly by processing and extracting features in the image, then detecting potential human joint points, and finally clustering and modeling the detected joint points. This paper introduces three methods based on coordinate regression method, heat map-based detection method and regression and detection mixed model, and presents a clear and clear presentation of the current research path of 2D human pose estimation.

關鍵字

2d ； human pose estimation ； CNN ； deep learning

參考文獻

LeCun Y, Boser B, Denker J S, et al. Backpropagation applied to handwritten zip code recognition [J]. Neural computation, 1989, 1(4): 541-551.

Girshick R.Fast R- CNN[C]//Proceedings of the IEEE International Conference on Computer Vision, 2015: 1440-1448.

Li S Z. Markov random field modeling in image analysis [M]. Springer Science & Business Media, 2009.

Wei S E, Ramakrishna V, Kanade T, et al. Convolutional pose machines[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 4724-4732.

Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation [C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 3431-3440.