filename     : Pra20a.pdf
entry        : inproceedings
conference   : Computer Vision and Pattern Recognition
pages        : 5861-5870
year         : 2020
month        : June
title        : Attention-Driven Cropping for Very High Resolution Facial Landmark Detection
subtitle     :
author       : Prashanth Chandran, Derek Bradley, Thabo Beeler, Markus Gross
booktitle    : Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
ISSN/ISBN    : 1558-0814
editor       : 
publisher    : IEEE
publ.place   :
volume       : 53
issue        : 10
language     : English
keywords     : Facial landmark detection, Attention, Keypoint detection, Pose estimation
abstract     : Facial landmark detection is a fundamental task for many consumer and high-end applications and is almost entirely solved by machine learning methods today. Existing datasets used to train such algorithms are primarily made up of only low resolution images, and current algorithms are limited to inputs of comparable quality and resolution as the training dataset. On the other hand, high resolution imagery is becoming increasingly more common as consumer cameras improve in quality every year. Therefore, there is need for algorithms that can leverage the rich information available in high resolution imagery. Naively attempting to reuse existing network architectures on high resolution imagery is prohibitive due to memory bottlenecks on GPUs. The only current solution is to downsample the images, sacrificing resolution and quality. Building on top of recent progress in attention-based networks, we present a novel, fully convolutional regional architecture that is specially designed for predicting landmarks on very high resolution facial images without downsampling. We demonstrate the flexibility of our architecture by training the proposed model with images of resolutions ranging from 256 x 256 to 4K. In addition to being the first method for facial landmark detection on high resolution images, our approach achieves superior performance over traditional (holistic) state-of-the-art architectures across ALL resolutions, leading to a general-purpose, extremely flexible, high quality landmark detector.