</> Code with Papers by Syntology

Datasets › COFW

COFW (Caltech Occluded Faces in the Wild)

Introduced in Robust Face Landmark Estimation under Occlusion1 Jan 2013 archive 2025-07-28

The Caltech Occluded Faces in the Wild (COFW) dataset is designed to present faces in real-world conditions. Faces show large variations in shape and occlusions due to differences in pose, expression, use of accessories such as sunglasses and hats and interactions with objects (e.g. food, hands, microphones,
etc.). All images were hand annotated using the same 29 landmarks as in LFPW. Both the landmark positions as well as their occluded/unoccluded state were annotated. The faces are occluded to different degrees, with large variations in the type of occlusions encountered. COFW has an average occlusion of over 23.

Source: http://www.vision.caltech.edu/xpburgos/ICCV13/#dataset Image Source: http://www.vision.caltech.edu/xpburgos/ICCV13/#dataset

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 5 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Face Alignment COFW FiFA NME (inter-ocular) 2.96 Fiducial Focus Augmentation for Facial Landmark Detection — 28 Compare
Face Alignment COFW-68 SPIGA NME (inter-ocular) 3.93 Shape Preserving Facial Landmarks with Graph Attention Networks andresprados/spiga 7 Compare
Face Alignment COFW-68 (300WLP) SH-FAN NME (box) 2.47 Subpixel Heatmap Regression for Facial Landmark Localization — 4 Compare
Facial Landmark Detection COFW D-ViT NME (inter-pupil) 4.13 Cascaded Dual Vision Transformer for Accurate Facial... Human3DAIGC/AccurateFacialLandmarkDetection 2 Compare
Head Pose Estimation COFW ASMNet MAE pitch (º) 2.72 ASMNet: a Lightweight Deep Neural Network for Face... aliprf/ASMNet 1 Compare

Papers archive 2025-07-28

30 shown of 34 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 114. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Cascaded Dual Vision Transformer for Accurate Facial Landmark Detection 1 1 8 Nov 2024 not harvested
Fiducial Focus Augmentation for Facial Landmark Detection 0 2 23 Feb 2024 not harvested
STAR Loss: Reducing Semantic Ambiguity in Facial Landmark Detection 1 1 5 Jun 2023 ran 4 of 4 samples (0 unverified; 4 pointer-only for licence)
Shape Preserving Facial Landmarks with Graph Attention Networks 1 1 13 Oct 2022 not harvested
Towards Accurate Facial Landmark Detection via Cascaded Transformers 0 1 23 Aug 2022 not harvested
ACR Loss: Adaptive Coordinate-based Regression Loss for Face Alignment 1 1 29 Mar 2022 not harvested
Sparse Local Patch Transformer for Robust Face Alignment and Landmarks Inherent Relation Learning 1 2 13 Mar 2022 not harvested
Occlusion-Robust Face Alignment Using a Viewpoint-Invariant Hierarchical Network Architecture 1 1 1 Jan 2022 not harvested
Facial Landmark Points Detection Using Knowledge Distillation-Based Neural Networks 1 2 13 Nov 2021 not harvested
Subpixel Heatmap Regression for Facial Landmark Localization 0 2 3 Nov 2021 not harvested
Revisiting Quantization Error in Face Alignment 0 1 13 Sep 2021 not harvested
When Liebig's Barrel Meets Facial Landmark Detection: A Practical Model 0 1 27 May 2021 not harvested
HIH: Towards More Accurate Face Alignment via Heatmap in Heatmap 1 1 7 Apr 2021 not harvested
Pre-training strategies and datasets for facial representation learning 2 1 30 Mar 2021 not harvested
ASMNet: a Lightweight Deep Neural Network for Face Alignment and Pose Estimation 1 1 27 Feb 2021 not harvested
Improving Robustness of Facial Landmark Detection by Defending Against Adversarial Attacks 1 1 1 Jan 2021 not harvested
Multi-task head pose estimation in-the-wild 1 2 22 Dec 2020 not harvested
ATF: Towards Robust Face Alignment via Leveraging Similarity and Diversity across Different Datasets 1 1 12 Oct 2020 not harvested
PropagationNet: Propagate Points to Curve to Learn Structure Information 0 1 25 Jun 2020 not harvested
Fast and Accurate: Structure Coherence Component for Face Alignment 0 1 21 Jun 2020 not harvested
Structured Landmark Detection via Topology-Adapting Deep Graph Learning 2 1 17 Apr 2020 not harvested
LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood 1 2 6 Apr 2020 not harvested
Pixel-in-Pixel Net: Towards Efficient Facial Landmark Detection in the Wild 2 1 8 Mar 2020 ran 2 of 17 samples (15 unverified)
Cascade of Encoder-Decoder CNNs with Learned Coordinates Regressor for Robust Facial Landmarks Detection 1 1 15 Oct 2019 not harvested
Face Alignment With Kernel Density Deep Neural Network 0 1 1 Oct 2019 not harvested
Deep High-Resolution Representation Learning for Visual Recognition 42 2 20 Aug 2019 ran 3 of 34 samples (31 unverified; 16 pointer-only for licence)
High-Resolution Representations for Labeling Pixels and Regions 39 1 9 Apr 2019 ran 3 of 18 samples (15 unverified; 5 pointer-only for licence)
Face Alignment using a 3D Deeply-initialized Ensemble of Regression Trees 1 1 5 Feb 2019 not harvested
Stacked Dense U-Nets with Dual Transformers for Robust Face Alignment 3 1 5 Dec 2018 not harvested
A Deeply-initialized Coarse-to-fine Ensemble of Regression Trees for Face Alignment 1 1 1 Sep 2018 not harvested

The full list of 34 is in the JSON twin.

Dataset loaders archive 2025-07-28

open-mmlab/mmposeopen-mmlab/mmposepytorch

1 loader as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

Face AlignmentFacial Landmark DetectionHead Pose Estimation

License archive 2025-07-28

Unknown

Modalities archive 2025-07-28

Images

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • COFW
  • COFW-68
  • COFW-68 (300WLP)

3 variant names, as the archive lists them.

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