Calibration free, user-independent gaze estimation with tensor analysis

作者：

Highlights：

• Tensor analysis framework for user-independent gaze detection without calibration

• Decomposes high dimension gaze data into factors selected by LASSO regression

• Evaluation includes individual, gaze, user-screen distances and session differences.

• Reduce performance gap between user-dependent and user-independent conditions

• Effective solution for gaze-aware multimodal interfaces

摘要

•Tensor analysis framework for user-independent gaze detection without calibration•Decomposes high dimension gaze data into factors selected by LASSO regression•Evaluation includes individual, gaze, user-screen distances and session differences.•Reduce performance gap between user-dependent and user-independent conditions•Effective solution for gaze-aware multimodal interfaces

论文关键词：User-independent gaze estimation,Tensor analysis,LASSO regression,Domain adaptation,Human computer interaction

论文评审过程：Received 11 April 2017, Revised 8 February 2018, Accepted 3 April 2018, Available online 17 April 2018, Version of Record 11 May 2018.

论文官网地址：https://doi.org/10.1016/j.imavis.2018.04.001