A lightweight 3D dense facial landmark estimation model from position map data

TL;DR

This paper introduces a lightweight 3D facial landmark estimation model using position map data, creating a dense keypoint dataset and demonstrating superior performance with minimal computational resources.

Contribution

The authors propose a novel pipeline to generate dense facial keypoints from position map data and train a compact MobileNet-based model for 3D landmark estimation.

Findings

Outperforms existing methods on 68 keypoint detection tasks

Effective in extreme head poses and occlusions

Low computational cost and small model size

Abstract

The incorporation of 3D data in facial analysis tasks has gained popularity in recent years. Though it provides a more accurate and detailed representation of the human face, accruing 3D face data is more complex and expensive than 2D face images. Either one has to rely on expensive 3D scanners or depth sensors which are prone to noise. An alternative option is the reconstruction of 3D faces from uncalibrated 2D images in an unsupervised way without any ground truth 3D data. However, such approaches are computationally expensive and the learned model size is not suitable for mobile or other edge device applications. Predicting dense 3D landmarks over the whole face can overcome this issue. As there is no public dataset available containing dense landmarks, we propose a pipeline to create a dense keypoint training dataset containing 520 key points across the whole face from an existing facial position map data. We train a lightweight MobileNet-based regressor model with the generated data. As we do not have access to any evaluation dataset with dense landmarks in it we evaluate our model against the 68 keypoint detection task. Experimental results show that our trained model outperforms many of the existing methods in spite of its lower model size and minimal computational cost. Also, the qualitative evaluation shows the efficiency of our trained models in extreme head pose angles as well as other facial variations and occlusions.