# Driver Gaze Zone Estimation using Convolutional Neural Networks: A General Framework and Ablative Analysis

arXiv: https://arxiv.org/abs/1802.02690 · HTML: https://mlsanity.com/paper/1802.02690/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Feb 2018.

## Built on by (most cited first, 20 of 20)

- [Automatic Gaze Analysis: A Survey of Deep Learning based Approaches](https://mlsanity.com/paper/2108.05479/) (2021)
- [Synthetic Distracted Driving (SynDD2) dataset for analyzing distracted behaviors and various gaze zones of a driver](https://mlsanity.com/paper/2204.08096/) (2022)
- [What Do You See in Vehicle? Comprehensive Vision Solution for In-Vehicle Gaze Estimation](https://mlsanity.com/paper/2403.15664/) (2024)
- [Autonomous Vehicles that Alert Humans to Take-Over Controls: Modeling with Real-World Data](https://mlsanity.com/paper/2104.11489/) (2021)
- [Safe Control Transitions: Machine Vision Based Observable Readiness Index and Data-Driven Takeover Time Prediction](https://mlsanity.com/paper/2301.05805/) (2023)
- [A Review of Driver Gaze Estimation and Application in Gaze Behavior Understanding](https://mlsanity.com/paper/2307.01470/) (2023)
- [Predicting Take-over Time for Autonomous Driving with Real-World Data: Robust Data Augmentation, Models, and Evaluation](https://mlsanity.com/paper/2107.12932/) (2021)
- [MTGLS: Multi-Task Gaze Estimation with Limited Supervision](https://mlsanity.com/paper/2110.12100/) (2021)
- [Behavioral Research and Practical Models of Drivers' Attention](https://mlsanity.com/paper/2104.05677/) (2021)
- [AV-Gaze: A Study on the Effectiveness of Audio Guided Visual Attention Estimation for Non-Profilic Faces](https://mlsanity.com/paper/2207.03048/) (2022)
- [Learning to Find Missing Video Frames with Synthetic Data Augmentation: A General Framework and Application in Generating Thermal Images Using RGB Cameras](https://mlsanity.com/paper/2403.00196/) (2024)
- [Driver Attention Tracking and Analysis](https://mlsanity.com/paper/2404.07122/) (2024)
- [HPE-CogVLM: Advancing Vision Language Models with a Head Pose Grounding Task](https://mlsanity.com/paper/2406.01914/) (2024)
- [EraW-Net: Enhance-Refine-Align W-Net for Scene-Associated Driver Attention Estimation](https://mlsanity.com/paper/2408.08570/) (2024)
- [SGAP-Gaze: Scene Grid Attention Based Point-of-Gaze Estimation Network for Driver Gaze](https://mlsanity.com/paper/2604.19888/) (2026)
- [Dense Residual Networks for Gaze Mapping on Indian Roads](https://mlsanity.com/paper/2203.11611/) (2022)
- [RAZE: Region Guided Self-Supervised Gaze Representation Learning](https://mlsanity.com/paper/2208.02485/) (2022)
- [Comparative Analysis of Kinect-Based and Oculus-Based Gaze Region Estimation Methods in a Driving Simulator](https://mlsanity.com/paper/2402.05248/) (2024)
- [Spatiotemporal Facial Action Unit Detection using Twin Cycle Autoencoders for Driver Monitoring](https://mlsanity.com/paper/2607.16760/) (2026)
- [TransGaze-Object: Transformer Based Driver Gaze Object Prediction Framework in Real Driving](https://mlsanity.com/paper/2609.10139/) (2026)

## Abstract (arXiv)

Driver gaze has been shown to be an excellent surrogate for driver attention in intelligent vehicles. With the recent surge of highly autonomous vehicles, driver gaze can be useful for determining the handoff time to a human driver. While there has been significant improvement in personalized driver gaze zone estimation systems, a generalized system which is invariant to different subjects, perspectives and scales is still lacking. We take a step towards this generalized system using Convolutional Neural Networks (CNNs). We finetune 4 popular CNN architectures for this task, and provide extensive comparisons of their outputs. We additionally experiment with different input image patches, and also examine how image size affects performance. For training and testing the networks, we collect a large naturalistic driving dataset comprising of 11 long drives, driven by 10 subjects in two different cars. Our best performing model achieves an accuracy of 95.18% during cross-subject testing, outperforming current state of the art techniques for this task. Finally, we evaluate our best performing model on the publicly available Columbia Gaze Dataset comprising of images from 56 subjects with varying head pose and gaze directions. Without any training, our model successfully encodes the different gaze directions on this diverse dataset, demonstrating good generalization capabilities.