Papers › On the power of data augmentation for head pose estimation

On the power of data augmentation for head pose estimation

7 Jul 2024arXiv:2407.05357archive 2025-07-28

Michael Welter

Deep learning has been impressively successful in the last decade in predicting human head poses from monocular images. However, for in-the-wild inputs the research community relies predominantly on a single training set, 300W-LP, of semisynthetic nature without many alternatives. This paper focuses on gradual extension and improvement of the data to explore the performance achievable with augmentation and synthesis strategies further. Modeling-wise a novel multitask head/loss design which includes uncertainty estimation is proposed. Overall, the thus obtained models are small, efficient, suitable for full 6 DoF pose estimation, and exhibit very competitive accuracy.

PaperPDFCode

Code

opentrack/neuralnet-tracker-traincode officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Data AugmentationFace AlignmentHead Pose EstimationPose Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Face Alignment AFLW2000-3D OpNet Balanced NME (2D Sparse Alignment) 3.55% #8 of 14 Archive leaderboard report
Head Pose Estimation AFLW2000 OpNet Geodesic Error (GE) 5.23 #1 of 25 Archive leaderboard report
Head Pose Estimation AFLW2000 OpNet MAE 3.15 #1 of 25 Archive leaderboard report
Head Pose Estimation BIWI OpNet Geodesic Error (GE) 7.01 #8 of 29 Archive leaderboard report
Head Pose Estimation BIWI OpNet Geodesic Error - aligned (GE) 4.72 #8 of 29 Archive leaderboard report
Head Pose Estimation BIWI OpNet MAE (trained with other data) 3.57 #8 of 29 Archive leaderboard report
Head Pose Estimation BIWI OpNet MAE-aligned (trained with other data) 2.65 #8 of 29 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

SET

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections