Face Alignment On 300W

Metrics

NME_inter-ocular (%, Challenge)

NME_inter-ocular (%, Common)

NME_inter-ocular (%, Full)

NME_inter-pupil (%, Challenge)

NME_inter-pupil (%, Common)

NME_inter-pupil (%, Full)

Results

Performance results of various models on this benchmark

Model Name	NME_inter-ocular (%, Challenge)	NME_inter-ocular (%, Common)	NME_inter-ocular (%, Full)	NME_inter-pupil (%, Challenge)	NME_inter-pupil (%, Common)	NME_inter-pupil (%, Full)	Paper Title	Repository
AWing	4.52	2.72	3.07	6.52	3.77	4.31	Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression
PropNet	3.99	2.67	2.93	5.75	3.7	4.1	PropagationNet: Propagate Points to Curve to Learn Structure Information	-
SLPT	4.90	2.75	3.17	-	-	-	Sparse Local Patch Transformer for Robust Face Alignment and Landmarks Inherent Relation Learning
GRegNet + LRefNet	4.78	2.71	3.12	6.89	3.76	4.37	Efficient and Accurate Face Alignment by Global Regression and Cascaded Local Refinement	-
LDDMM-Face	5.4	3.07	3.53	-	-	-	LDDMM-Face: Large Deformation Diffeomorphic Metric Learning for Flexible and Consistent Face Alignment	-
DAN-Menpo	4.88	3.09	3.44	7.05	4.29	4.83	Deep Alignment Network: A convolutional neural network for robust face alignment
ODN	6.67	3.56	4.17	-	-	-	Robust Facial Landmark Detection via Occlusion-Adaptive Deep Networks	-
AVS	6.49	3.21	3.86	-	-	-	Aggregation via Separation: Boosting Facial Landmark Detector with Semi-Supervised Style Translation	-
ATF	4.89	2.75	3.17	-	-	-	ATF: Towards Robust Face Alignment via Leveraging Similarity and Diversity across Different Datasets	-
CHR2C	5.15	2.85	3.3	7.44	3.96	4.64	Cascade of Encoder-Decoder CNNs with Learned Coordinates Regressor for Robust Facial Landmarks Detection
HIH	4.89	2.65	3.09	-	-	-	HIH: Towards More Accurate Face Alignment via Heatmap in Heatmap
Wing	-	-	-	7.18	3.27	4.04	Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks
EF-3ACR	5.36	3.36	3.75	-	-	-	ACR Loss: Adaptive Coordinate-based Regression Loss for Face Alignment	-
SAN	6.60	3.34	3.98	-	-	-	Style Aggregated Network for Facial Landmark Detection
DenseLandmarks (GNLL)	4.8	3.03	-	-	-	-	3D face reconstruction with dense landmarks	-
FaRL-B (epoch 16)	4.45	2.56	2.93	6.42	3.53	4.11	General Facial Representation Learning in a Visual-Linguistic Manner
MobileNetV2	7.35	3.88	4.59	-	-	-	ASMNet: a Lightweight Deep Neural Network for Face Alignment and Pose Estimation	-
HRNet	5.15	2.87	3.32	-	-	-	Deep High-Resolution Representation Learning for Visual Recognition
3FabRec	5.74	3.36	3.82	-	-	-	3FabRec: Fast Few-shot Face alignment by Reconstruction
SCC	4.93	2.88	3.28	-	-	-	Fast and Accurate: Structure Coherence Component for Face Alignment	-

0 of 48 row(s) selected.