Papers › Cascade of Encoder-Decoder CNNs with Learned Coordinates Regressor for Robust Facial...
Cascade of Encoder-Decoder CNNs with Learned Coordinates Regressor for Robust Facial Landmarks Detection
Roberto Valle, Jose M. Buenaposada, Luis Baumela
Convolutional Neural Nets (CNNs) have become the reference technology for many computer vision problems. Although CNNs for facial landmark detection are very robust, they still lack accuracy when processing images acquired in unrestricted conditions. In this paper we investigate the use of a cascade of Neural Net regressors to increase the accuracy of the estimated facial landmarks. To this end we append two encoder-decoder CNNs with the same architecture. The first net produces a set of heatmaps with a rough estimation of landmark locations. The second, trained with synthetically generated occlusions, refines the location of ambiguous and occluded landmarks. Finally, a densely connected layer with shared weights among all heatmaps, accurately regresses the landmark coordinates. The proposed approach achieves state-of-the-art results in 300W, COFW and WFLW that are widely considered the most challenging public data sets.
Code
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
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Face Alignment | 300W | CHR2C | NME_inter-ocular (%, Challenge) | 5.15 | #22 of 48 | Archive leaderboard | report |
| Face Alignment | 300W | CHR2C | NME_inter-ocular (%, Common) | 2.85 | #22 of 48 | Archive leaderboard | report |
| Face Alignment | 300W | CHR2C | NME_inter-ocular (%, Full) | 3.3 | #22 of 48 | Archive leaderboard | report |
| Face Alignment | 300W | CHR2C | NME_inter-pupil (%, Challenge) | 7.44 | #22 of 48 | Archive leaderboard | report |
| Face Alignment | 300W | CHR2C | NME_inter-pupil (%, Common) | 3.96 | #22 of 48 | Archive leaderboard | report |
| Face Alignment | 300W | CHR2C | NME_inter-pupil (%, Full) | 4.64 | #22 of 48 | Archive leaderboard | report |
| Face Alignment | COFW | CHR2C (Inter-pupils Norm) | NME (inter-pupil) | 5.09% | #25 of 28 | Archive leaderboard | report |
| Face Alignment | WFLW | CHR2C | AUC@10 (inter-ocular) | 57.55 | #19 of 36 | Archive leaderboard | report |
| Face Alignment | WFLW | CHR2C | FR@10 (inter-ocular) | 3.55 | #19 of 36 | Archive leaderboard | report |
| Face Alignment | WFLW | CHR2C | NME (inter-ocular) | 4.39 | #19 of 36 | Archive leaderboard | report |
| Facial Landmark Detection | 300W | CHR2C (Inter-ocular Norm) | NME | 3.3 | #7 of 15 | 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