Landmark Facial Detection by using Gaussian Regression Guided Network

  • Shubham Mahajan, Suvajit Lodh, Anil Kumar Bhardwaj, Amit Kant Pandit


The areas of the fiducial facial milestone focuses around facial components and facial contour capture their gidandnon - unbending facial deformations due to head movements and facial expressions. They are hence important for various facial analysis tasks. Numerous facial landmark identification calculations have been created to naturally identify those key focuses throughout the years. Facial landmark recognition is significant earlier data for other face arrangement issues, for example, head present estimation, facial feeling demeanor, and face displaying. Among the run of the mill facial landmark identification calculations, TREE [1] calculation using fell regression strategy distinguish facial landmark quicker than different calculations utilizing little preparing information. Work done in paper, we separate the Gaussian guided landmark map from the TREE and foresee remaining facial landmark spots by means of landmark regression network. By utilizing the Gaussian landmark highlight map as earlier data, the regression arrange adequately manages the bearing of the refined landmark spots on the Helen examination dataset.

Keywords: Face alignment, landmark and feature point detection, gaussian regression.

How to Cite
Amit Kant Pandit, S. M. S. L. A. K. B. (2019). Landmark Facial Detection by using Gaussian Regression Guided Network. International Journal of Control and Automation, 12(5), 431 - 436. Retrieved from