Expression robust 3D face landmarking using thresholded surface normals
作者：
Highlights：
• A novel 3D facial landmarking algorithm based on thresholded surface normals maps is proposed.
• The potential of using surface normals maps for feature detection is explored.
• Using thresholded surface normals can help facial roll and yaw rotation calibration.
• Seven facial landmarks (the tip, root, subnasal, alar grooves and eye corners) are robustly localized on the well-aligned 3D face.
• Results on the Bosphorus, FRGC and BU-3DFE databases show that the detected landmarks possess high within-class consistency and accuracy under different expressions.
摘要
•A novel 3D facial landmarking algorithm based on thresholded surface normals maps is proposed.•The potential of using surface normals maps for feature detection is explored.•Using thresholded surface normals can help facial roll and yaw rotation calibration.•Seven facial landmarks (the tip, root, subnasal, alar grooves and eye corners) are robustly localized on the well-aligned 3D face.•Results on the Bosphorus, FRGC and BU-3DFE databases show that the detected landmarks possess high within-class consistency and accuracy under different expressions.
论文关键词：3D face landmarking,Surface normals
论文评审过程：Received 13 June 2017, Revised 28 October 2017, Accepted 14 January 2018, Available online 17 January 2018, Version of Record 3 February 2018.
论文官网地址：https://doi.org/10.1016/j.patcog.2018.01.011
