# RealHePoNet: a robust single-stage ConvNet for head pose estimation in the wild

arXiv: https://arxiv.org/abs/2011.01890 · HTML: https://mlsanity.com/paper/2011.01890/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Nov 2020.

## Built on by (most cited first, 7 of 7)

- [edBB-Demo: Biometrics and Behavior Analysis for Online Educational Platforms](https://mlsanity.com/paper/2211.09210/) (2022)
- [A multimodal dataset for understanding the impact of mobile phones on remote online virtual education](https://mlsanity.com/paper/2412.14195/) (2024)
- [Biometrics and Behavior Analysis for Detecting Distractions in e-Learning](https://mlsanity.com/paper/2405.15434/) (2024)
- [MATT: Multimodal Attention Level Estimation for e-learning Platforms](https://mlsanity.com/paper/2301.09174/) (2023)
- [DeepFace-Attention: Multimodal Face Biometrics for Attention Estimation with Application to e-Learning](https://mlsanity.com/paper/2408.05523/) (2024)
- [A Multimodal Dataset of Student Oral Presentations with Sensors and Evaluation Data](https://mlsanity.com/paper/2601.07576/) (2026)
- [Body and Head Orientation Estimation from Low-Resolution Point Clouds in Surveillance Settings](https://mlsanity.com/paper/2311.00343/) (2023)

## Abstract (arXiv)

Human head pose estimation in images has applications in many fields such as human-computer interaction or video surveillance tasks. In this work, we address this problem, defined here as the estimation of both vertical (tilt/pitch) and horizontal (pan/yaw) angles, through the use of a single Convolutional Neural Network (ConvNet) model, trying to balance precision and inference speed in order to maximize its usability in real-world applications. Our model is trained over the combination of two datasets: 'Pointing'04' (aiming at covering a wide range of poses) and 'Annotated Facial Landmarks in the Wild' (in order to improve robustness of our model for its use on real-world images). Three different partitions of the combined dataset are defined and used for training, validation and testing purposes. As a result of this work, we have obtained a trained ConvNet model, coined RealHePoNet, that given a low-resolution grayscale input image, and without the need of using facial landmarks, is able to estimate with low error both tilt and pan angles (~4.4deg average error on the test partition). Also, given its low inference time (~6 ms per head), we consider our model usable even when paired with medium-spec hardware (i.e. GTX 1060 GPU). * Code available at: https://github.com/rafabs97/headposefinal * Demo video at: https://www.youtube.com/watch?v=2UeuXh5DjAE