Papers › Face Alignment using a 3D Deeply-initialized Ensemble of Regression Trees

Face Alignment using a 3D Deeply-initialized Ensemble of Regression Trees

5 Feb 2019arXiv:1902.01831archive 2025-07-28

Roberto Valle, José M. Buenaposada, Antonio Valdés, Luis Baumela

Face alignment algorithms locate a set of landmark points in images of faces taken in unrestricted situations. State-of-the-art approaches typically fail or lose accuracy in the presence of occlusions, strong deformations, large pose variations and ambiguous configurations. In this paper we present 3DDE, a robust and efficient face alignment algorithm based on a coarse-to-fine cascade of ensembles of regression trees. It is initialized by robustly fitting a 3D face model to the probability maps produced by a convolutional neural network. With this initialization we address self-occlusions and large face rotations. Further, the regressor implicitly imposes a prior face shape on the solution, addressing occlusions and ambiguous face configurations. Its coarse-to-fine structure tackles the combinatorial explosion of parts deformation. In the experiments performed, 3DDE improves the state-of-the-art in 300W, COFW, AFLW and WFLW data sets. Finally, we perform cross-dataset experiments that reveal the existence of a significant data set bias in these benchmarks.

PaperPDFCode

Code

bobetocalo/bobetocalo_eccv18 mentioned on GitHubtf 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

Face AlignmentFace ModelFacial Landmark Detectionregression

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Face Alignment 300W 3DDE NME_inter-ocular (%, Challenge) 4.92 #15 of 48 Archive leaderboard report
Face Alignment 300W 3DDE NME_inter-ocular (%, Common) 2.69 #15 of 48 Archive leaderboard report
Face Alignment 300W 3DDE NME_inter-ocular (%, Full) 3.13 #15 of 48 Archive leaderboard report
Face Alignment 300W 3DDE NME_inter-pupil (%, Challenge) 7.10 #15 of 48 Archive leaderboard report
Face Alignment 300W 3DDE NME_inter-pupil (%, Common) 3.73 #15 of 48 Archive leaderboard report
Face Alignment 300W 3DDE NME_inter-pupil (%, Full) 4.39 #15 of 48 Archive leaderboard report
Face Alignment 300W Split 2 3DDE AUC@8 (inter-ocular) 53.94 #5 of 7 Archive leaderboard report
Face Alignment 300W Split 2 3DDE FR@8 (inter-ocular) 2.33 #5 of 7 Archive leaderboard report
Face Alignment 300W Split 2 3DDE NME (inter-ocular) 3.73 #5 of 7 Archive leaderboard report
Face Alignment COFW 3DDE (Inter-pupil Norm) NME (inter-pupil) 5.11% #24 of 28 Archive leaderboard report
Face Alignment COFW 3DDE (Inter-pupil Norm) Recall at 80% precision (Landmarks Visibility) 63.89 #24 of 28 Archive leaderboard report
Face Alignment WFLW 3DDE AUC@10 (inter-ocular) 55.44 #26 of 36 Archive leaderboard report
Face Alignment WFLW 3DDE FR@10 (inter-ocular) 5.04 #26 of 36 Archive leaderboard report
Face Alignment WFLW 3DDE NME (inter-ocular) 4.68 #26 of 36 Archive leaderboard report
Facial Landmark Detection 300W 3DDE (Inter-ocular Norm) NME 3.13 #5 of 15 Archive leaderboard report
Facial Landmark Detection AFLW-Full 3DDE (Box height Norm, 19 landmarks - no earlobs) Mean NME 2.01 #5 of 5 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.

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