Training data

The dataset is an incremental of JD-landmark [1-7], consisting of about 20,000 images with 106-point landmark annotation. Different to JD-landmark, all of the incremental samples are real-world data with human annotation, without any algo-generated ones. The images and landmarks will be released on March 30, 2020.

Validation data

It consists of 2,000 images, which are selected from open source web face dataset [8] and cover large variations of pose, expression and occlusion. The images without ground truth will be released on April 7, 2020.

Test data

It contains 2,000 web face images as well. To prevent cheating on the test set, the test images without ground truth will be released on June 24, 2020 [00:00 a.m. UTC+1], and the submission deadline is June 25, 2020 [00:00 a.m. UTC+1]. Participants will have 24 hours to evaluate on the test set.

References

[1] Y. Liu, H. Shen, Y. Si, X. Wang, X. Zhu, H. Shi et al. Grand challenge of 106-point facial landmark localization. In ICMEW, 2019.

[2] C. Sagonas, G. Tzimiropoulos, S. Zafeiriou, and M. Pantic. 300 faces in-the-wild challenge: The first facial landmark localization Challenge. In ICCVW, 2013.

[3] C. Sagonas, E. Antonakos, G. Tzimiropoulos, S. Zafeiriou, and M. Pantic. 300 faces in-the-wild challenge: Database and results. In IVC, 2016.

[4] Belhumeur, P., Jacobs, D., Kriegman, D., Kumar, N. Localizing parts of faces using a consensus of exemplars. In Computer Vision and Pattern Recognition. In CVPR, 2011.

[5] X. Zhu, D. Ramanan. Face detection, pose estimation and landmark localization in the wild. In CVPR, 2012.

[6] Vuong Le, Jonathan Brandt, Zhe Lin, Lubomir Boudev, Thomas S. Huang. Interactive facial feature localization. In ECCV, 2012.

[7] C. Sagonas, G. Tzimiropoulos, S. Zafeiriou, and M. Pantic. A semi-automatic methodology for facial landmark annotation. In CVPR, 2013.

[8] I. Kemelmacher-Shlizerman, S. M. Seitz, D. Miller, and E. Brossard. The megaface benchmark: 1 million faces for recognition at scale. In CVPR, 2016.

Additional Information

• The dataset is available for non-commercial research purposes only.
• You agree not to reproduce, duplicate, copy, sell, trade, resell or exploit for any commercial purposes, any portion of the images and any portion of derived data.
• You agree not to further copy, publish or distribute any portion of annotations of the dataset. Except, for internal use at a single site within the same organization it is allowed to make copies of the dataset.
• We reserve the right to terminate your access to the dataset at any time.