AFLW2000-3D

Dataset Information
Modalities
Images, 3D
Introduced
2016
License
Unknown
Homepage

Overview

AFLW2000-3D is a dataset of 2000 images that have been annotated with image-level 68-point 3D facial landmarks. This dataset is used for evaluation of 3D facial landmark detection models. The head poses are very diverse and often hard to be detected by a CNN-based face detector.

Source: https://www.tensorflow.org/datasets/catalog/aflw2k3d
Image Source: http://www.cbsr.ia.ac.cn/users/xiangyuzhu/projects/3DDFA/main.htm

Variants: AFLW2000, AFLW2000-3D

Associated Benchmarks

This dataset is used in 2 benchmarks:

Recent Benchmark Submissions

Task Model Paper Date
3D Face Reconstruction DSFNet-is DSFNet: Dual Space Fusion Network … 2023-05-19
3D Face Reconstruction SADRNet SADRNet: Self-Aligned Dual Face Regression … 2021-06-06
3D Face Reconstruction B-spline FFD Learning Free-Form Deformation for 3D … 2021-05-31
3D Face Reconstruction VGG-F Pre-training strategies and datasets for … 2021-03-30
3D Face Reconstruction 3DDFA-V2 Towards Fast, Accurate and Stable … 2020-09-21
3D Face Reconstruction PRN Joint 3D Face Reconstruction and … 2018-03-21
Facial Landmark Detection JVCR Joint Voxel and Coordinate Regression … 2018-01-28
3D Face Reconstruction DeFA Dense Face Alignment 2017-09-05
3D Face Reconstruction 3DDFA Face Alignment Across Large Poses: … 2015-11-23

Research Papers

Recent papers with results on this dataset: