1. [Publications](/publications) 2. Few-Shot Adaptive Gaze Estimation # Few-Shot Adaptive Gaze Estimation ![](/sites/default/files/publications/teaser_0.gif) Inter-personal anatomical differences limit the accuracy of person-independent gaze estimation networks. Yet there is a need to lower gaze errors further to enable applications requiring higher quality. Further gains can be achieved by personalizing gaze networks, ideally with few calibration samples. However, over-parameterized neural networks are not amenable to learning from few examples as they can quickly over-fit. We embrace these challenges and propose a novel framework for Few-shot Adaptive GaZE Estimation (FAZE) for learning person-specific gaze networks with very few ( 9) calibration samples. FAZE learns a rotation aware latent representation of gaze via a disentangling encoder-decoder architecture along with a highly adaptable gaze estimator trained using meta-learning. It is capable of adapting to any new person to yield significant performance gains with as few as 3 samples, yielding state-of-the art performance of 3:18 on GazeCapture, a 19% improvement over prior art. We open-source our code at [https://github.com/NVlabs/few\_shot\_gaze](https://github.com/NVlabs/few_shot_gaze). ## Authors Seonwook Park (ETH Zürich) [Shalini De Mello](/person/shalini-de-mello) [Pavlo Molchanov](/person/pavlo-molchanov) [Umar Iqbal](/person/umar-iqbal) Otmar Hilliges (ETH Zürich) [Jan Kautz](/person/jan-kautz) ## Publication Date Tuesday, October 29, 2019 ## Published in [International Conference on Computer Vision (ICCV) 2019](http://iccv2019.thecvf.com/program/overview) ## Research Area [Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence) [Computer Vision](/research-area/computer-vision) [Human Computer Interaction](/research-area/human-computer-interaction) ## External Links [Code](https://github.com/NVlabs/few_shot_gaze) [ICCV 2019 Oral Presentation](https://conftube.com/video/ByfFufRhuRc?tocitem=17) [ETH Project Page](https://ait.ethz.ch/projects/2019/faze/) ## Uploaded Files [Paper](https://research.nvidia.com/sites/default/files/pubs/2019-10_Few-Shot-Adaptive-Gaze//1112.pdf "Open file in new window")2.28 MB [Supplementary](https://research.nvidia.com/sites/default/files/pubs/2019-10_Few-Shot-Adaptive-Gaze/1112-supp.pdf "Open file in new window")385.88 KB ## Award Oral ## Copyright This material is posted here with permission of the IEEE. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to .