pith. sign in

arxiv: 2606.20823 · v1 · pith:B77QDTB7new · submitted 2026-06-18 · 💻 cs.CV

NeoLoc-68: End-to-end 68-point neonatal facial landmark localisation in neonatal clinical environments

Abdullah Bin-Obaid , Maria M. Cobo , Rebeccah Slater , Lionel Tarassenko , Mauricio Villarroel This is my paper

Pith reviewed 2026-06-26 18:11 UTC · model grok-4.3

classification 💻 cs.CV
keywords facial landmark detectionneonatal facesclinical environments68-point landmarksYOLO keypoint modelpain assessmentgeneralizationend-to-end detection
0
0 comments X

The pith

Mixing standardized adult face images with neonatal frames trains the first end-to-end model to locate 68 landmarks on babies in clinical settings.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a detector that places 68 specific points on newborn faces even when the infants are in hospital environments surrounded by equipment and showing sudden movements. It standardizes over 37,000 images from public adult datasets to a common landmark scheme and combines them with over 1,000 manually labeled neonatal frames to train a regression model. The system starts from weights of a neonatal face detector and uses a keypoint approach to output the points directly. This step matters because accurate landmarks form the basis for automated systems that read facial expressions to judge pain without touching the baby. If the approach holds, it provides a foundation for contact-free monitoring tools that clinicians could use in neonatal care.

Core claim

The paper claims that standardizing 37,459 single-face images from 11 public datasets to a 68-point markup and mixing them with 1,123 annotated neonatal frames produces a training set that lets an adapted YOLO-based keypoint model regress 68 landmarks on neonatal faces in clinical conditions, delivering the lowest detection failure rate among tested baselines on a clinical test set before fine-tuning and further gains after fine-tuning, establishing the first end-to-end 68-point neonatal model.

What carries the argument

The YOLO-based keypoint model adapted to regress 68 facial landmarks, initialized with weights from a pretrained neonatal face detector.

If this is right

  • The model reaches state-of-the-art numbers on public face datasets using the standardized 68-point markup.
  • It generalizes to clinical neonatal images with the lowest failure rate among baselines even before any neonatal fine-tuning.
  • Fine-tuning on additional neonatal frames further lowers error rates and failure rates.
  • The outputs can feed directly into downstream neonatal health monitoring and facial expression analysis tasks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar mixing of public adult data with small specialized sets could reduce the need for large new annotations in other medical imaging domains with limited samples.
  • The landmark outputs could support video-based tracking of expression changes over time rather than single frames.
  • Deployment in neonatal units might allow continuous non-contact observation of distress signals during routine care.

Load-bearing premise

That standardizing adult images to a shared 68-point scheme and mixing them with limited neonatal frames creates a training distribution whose gap to real clinical neonatal images remains small enough for the observed generalization to hold.

What would settle it

A fresh clinical neonatal test set annotated by separate experts on which the model shows detection failure rates substantially above the reported levels would show the generalization claim does not hold.

Figures

Figures reproduced from arXiv: 2606.20823 by Abdullah Bin-Obaid, Lionel Tarassenko, Maria M. Cobo, Mauricio Villarroel, Rebeccah Slater.

Figure 1
Figure 1. Figure 1: The 68-point facial landmark annotation scheme and example annotations. The 68-point facial landmark configuration and an example annotation. (a) The landmarks are numbered in order and span the key facial regions, including the jawline, eyebrows, eyes, nose, and mouth (image adapted from [17]). (b) Example image showing landmark annota￾tions (green), the landmark-derived bounding box (red), the expanded g… view at source ↗
Figure 2
Figure 2. Figure 2: Example images from the selected facial landmark datasets used in this work. Each image is accompanied by its dataset name on the left and is shown with its corresponding set of 68 annotated facial landmarks, reformatted where necessary to follow the iBUG markup scheme. Images have been cropped around the face region to provide consistent framing and improve visual comparability across datasets. 9 [PITH_F… view at source ↗
Figure 3
Figure 3. Figure 3: Illustration of the weight transfer process from the detection model to the key￾points model. The high-level architectures of both models: the face detection model (top) and the keypoints model (bottom). Both share the same backbone (dark blue rectangle), neck (light blue rectangle), and detection heads (green rectangles). The keypoints model includes additional regressor heads (gold rectangles) for keypoi… view at source ↗
read the original abstract

Facial landmark localisation is a prerequisite for developing automated, non-contact neonatal pain assessment methods. Clinicians use pain scales to judge the severity of pain, many of which rely on facial expression. However, facial landmark detectors trained on adult faces perform poorly in neonatal clinical environments due to frequent occlusions caused by medical equipment, varied head poses, and challenging imaging conditions, including motion blur triggered by sudden pain-related movements. We propose an end-to-end facial landmark detector capable of predicting 68 landmarks on neonatal faces in clinical environments. We combined 37,459 single-face images from 11 public datasets, standardised to 68-point markup, with 1,123 manually annotated frames from a neonatal research dataset (totalling over 76,000 landmarks). A YOLO-based keypoint model was adapted to regress the facial landmarks, initialised with weights from a pretrained neonatal face detector. On public datasets, our proposed model achieved state-of-the-art performance: Normalised Mean Error (NME) = 5.37, Failure Rate (FR) = 12.5%, Area Under the Cumulative Error Curve (AUC) at AUC0.08 = 38.00% and AUC0.1 = 48.70%. On the clinical neonatal test set, before fine-tuning, the model achieved the lowest Detection Failure Rate (DFR) = 5.3% among all baselines and showed strong generalisation. After fine-tuning, performance improved further to NME = 6.36, FR = 22.30%, DFR = 1.77%, AUC0.08 = 29.24% and AUC0.1 = 40.25%. To the best of our knowledge, this represents the first end-to-end 68-point neonatal facial landmark detection model. With further dataset expansion and refinement, it could support downstream tasks in neonatal health monitoring and pain-related facial analysis.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces NeoLoc-68, claimed as the first end-to-end 68-point facial landmark detector for neonatal faces in clinical environments. It trains a YOLO-based keypoint regressor on 37,459 images from 11 public (mostly adult) datasets standardized to 68-point markup plus 1,123 manually annotated neonatal frames, initialized from pretrained neonatal face detector weights. Reported results include SOTA on public datasets (NME=5.37, FR=12.5%, AUC0.08=38.00%) and strong generalization on a held-out clinical neonatal test set (DFR=5.3% before fine-tuning; after fine-tuning NME=6.36, FR=22.30%, DFR=1.77%).

Significance. If the generalization results hold after addressing verification gaps, the work would be significant for enabling non-contact neonatal pain assessment via facial expression analysis under real clinical conditions (occlusions, motion blur, varied poses). The mixed-dataset training strategy and multi-metric evaluation (including DFR and AUC) are positive aspects; however, the absence of protocol details limits immediate utility for downstream tasks.

major comments (2)
  1. [Data section] Data section: The standardization of 37,459 images from 11 heterogeneous public datasets to a common 68-point markup is described at a high level but supplies no explicit landmark mapping, inter-annotator agreement, or geometric consistency checks against neonatal anatomy. This is load-bearing for the central generalization claim (DFR=5.3% on unseen clinical neonatal frames before fine-tuning), as systematic offsets from adult-to-68 conversion could be absorbed into the regressor without true domain transfer.
  2. [Experiments section] Experiments section: The manuscript reports numeric performance metrics, SOTA claims, and baseline comparisons but provides no experimental protocol, train-test split information, baseline implementation details, or error analysis. This prevents verification of the reported results (e.g., pre-fine-tune DFR=5.3% and post-fine-tune metrics) and undermines the soundness of the generalization statements.
minor comments (1)
  1. [Abstract] Abstract: The parenthetical '(totalling over 76,000 landmarks)' appears to count only the 1,123 neonatal frames (≈76k landmarks) while omitting the contribution from the 37,459 public images; this is a minor arithmetic/presentation inconsistency.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments, which highlight important areas for improving the clarity and verifiability of our work. We address each major comment point-by-point below and commit to revisions that will strengthen the manuscript without altering its core claims.

read point-by-point responses

  1. Referee: [Data section] The standardization of 37,459 images from 11 heterogeneous public datasets to a common 68-point markup is described at a high level but supplies no explicit landmark mapping, inter-annotator agreement, or geometric consistency checks against neonatal anatomy. This is load-bearing for the central generalization claim (DFR=5.3% on unseen clinical neonatal frames before fine-tuning), as systematic offsets from adult-to-68 conversion could be absorbed into the regressor without true domain transfer.

    Authors: We acknowledge that the Data section currently provides only a high-level description of the standardization process. To directly address this concern and support the generalization results, the revised manuscript will include an expanded Data section with explicit per-dataset landmark mappings to the 68-point format, available inter-annotator agreement statistics for the manually annotated neonatal frames, and a description of the geometric consistency checks applied during standardization to align with neonatal anatomy. These additions will clarify that the reported pre-fine-tuning DFR reflects genuine domain transfer rather than absorbed offsets. revision: yes

  2. Referee: [Experiments section] The manuscript reports numeric performance metrics, SOTA claims, and baseline comparisons but provides no experimental protocol, train-test split information, baseline implementation details, or error analysis. This prevents verification of the reported results (e.g., pre-fine-tune DFR=5.3% and post-fine-tune metrics) and undermines the soundness of the generalization statements.

    Authors: We agree that the absence of detailed experimental protocol information limits independent verification. In the revised manuscript, we will add a new Experimental Protocol subsection that specifies the train-test splits used for the public datasets and the held-out neonatal test set, full implementation details and hyperparameters for all baselines, the exact training procedure including initialization from the neonatal face detector, and an error analysis with qualitative examples of failure cases. This will enable full reproduction and verification of all reported metrics, including the pre- and post-fine-tuning DFR, NME, FR, and AUC values. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical evaluation on held-out sets

full rationale

The paper reports an ML training and evaluation pipeline for a YOLO-based keypoint regressor on a mixed adult+neonatal dataset, with performance measured via standard metrics (NME, FR, DFR, AUC) on explicitly held-out clinical neonatal test frames. No equations, derivations, or 'predictions' appear that reduce to fitted parameters or self-definitions by construction. No load-bearing self-citations or uniqueness theorems are invoked. The central claims rest on external test-set results rather than any internal redefinition or renaming of inputs, satisfying the self-contained criterion.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Central claim rests on transfer from mixed public datasets to neonatal clinical images; specific free parameters and assumptions are not detailed beyond the dataset counts and initialization choice.

free parameters (1)
  • pretrained neonatal face detector weights
    Model is initialised with weights from a pretrained neonatal face detector; this choice directly affects starting point and reported generalization.
axioms (1)
  • domain assumption Public adult face datasets can be standardized to the same 68-point markup and combined with neonatal frames without introducing systematic label noise or domain mismatch that invalidates clinical generalization
    The paper combines 37,459 images from 11 public datasets standardised to 68-point markup with 1,123 neonatal frames.

pith-pipeline@v0.9.1-grok · 5905 in / 1296 out tokens · 38022 ms · 2026-06-26T18:11:20.107289+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

39 extracted references · 10 canonical work pages

  1. [1]

    Facial landmark detection: A literature survey

    Yue Wu and Qiang Ji. Facial landmark detection: A literature survey. International Journal of Computer Vision, 127(2):115–142, Feb 2019. doi: 10.1007/s11263-018-1097-z

    work page doi:10.1007/s11263-018-1097-z 2019

  2. [2]

    InfanFace: Bridging the infant–adult domain gap in facial landmark estimation in the wild

    Meng Wan, Shu Zhu, Lei Luan, Gaurav Prateek, Xiaoming Huang, Rebecca Schwartz- Mette, Michelle Hayes, Eugene Zimmerman, and Sarah Ostadabbas. InfanFace: Bridging the infant–adult domain gap in facial landmark estimation in the wild. In Proceedings of the 26th International Conference on Pattern Recognition (ICPR), pages 4486–4492. IEEE, 2022

    2022

  3. [3]

    Ashraf, Melissa E

    Azadeh Asgarian, Shaundra Zhao, Afsaneh B. Ashraf, Melissa E. Browne, Kenneth M. Prkachin, Alex Mihailidis, Thomas Hadjistavropoulos, and Babak Taati. Limitations and biases in facial landmark detection: An empirical study on older adults with dementia. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW),...

    2019

  4. [4]

    Donald F. Huelke. An overview of anatomical considerations of infants and children in the adult world of automobile safety design. In Annual Proceedings of the Association for the Advancement of Automotive Medicine, volume 42, pages 93–113, 1998

    1998

  5. [5]

    Du- mont, and Farzaneh Marzbanrad

    Evan Grooby, Chhama Sitaula, Sogand Ahani, Liisa Holsti, Ayush Malhotra, Guy A. Du- mont, and Farzaneh Marzbanrad. Neonatal face and facial landmark detection from video recordings. In 2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) , pages 1–5. IEEE, Jul 2023

    2023

  6. [6]

    Fast facial landmark detection and appli- cations: A survey

    Kostiantyn Khabarlak and Larysa Koriashkina. Fast facial landmark detection and appli- cations: A survey. Journal of Computer Science and Technology , 22:e02, Apr 2022. URL https://journal.info.unlp.edu.ar/JCST/article/view/1972

    2022

  7. [7]

    One millisecond face alignment with an ensemble of regression trees

    Vahid Kazemi and Josephine Sullivan. One millisecond face alignment with an ensemble of regression trees. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 1867–1874, 2014. doi: 10.1109/CVPR.2014.241

    work page doi:10.1109/cvpr.2014.241 2014

  8. [8]

    Davis E. King. Dlib-ml: A machine learning toolkit. Journal of Machine Learning Research, 10:1755–1760, Dec 2009

    2009

  9. [9]

    How far are we from solving the 2d & 3d face alignment problem? (and a dataset of 230,000 3d facial landmarks)

    Adrian Bulat and Georgios Tzimiropoulos. How far are we from solving the 2d & 3d face alignment problem? (and a dataset of 230,000 3d facial landmarks). In Proceedings of the IEEE International Conference on Computer Vision (ICCV) , pages 1021–1030, 2017

    2017

  10. [10]

    Stacked hourglass networks for human pose estimation

    Alejandro Newell, Kaiyu Yang, and Jia Deng. Stacked hourglass networks for human pose estimation. In European Conference on Computer Vision (ECCV) , pages 483–499, Cham, Sep 2016. Springer International Publishing. 26

    2016

  11. [11]

    Binarized convolutional landmark localizers for human pose estimation and face alignment with limited resources

    Adrian Bulat and Georgios Tzimiropoulos. Binarized convolutional landmark localizers for human pose estimation and face alignment with limited resources. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) , pages 3706–3714, 2017

    2017

  12. [12]

    Pixel-in-pixel net: Towards efficient facial land- mark detection in the wild.International Journal of Computer Vision, 129(12):3174–3194, Dec 2021

    Haibo Jin, Shengcai Liao, and Ling Shao. Pixel-in-pixel net: Towards efficient facial land- mark detection in the wild.International Journal of Computer Vision, 129(12):3174–3194, Dec 2021. doi: 10.1007/s11263-021-01500-9

    work page doi:10.1007/s11263-021-01500-9 2021

  13. [13]

    Buenaposada, and Luis Baumela

    Antonio Prados-Torreblanca, José M. Buenaposada, and Luis Baumela. Shape preserving facial landmarks with graph attention networks, 2022. URL https://arxiv.org/abs/ 2210.07233. [Preprint]. Posted 2022 Oct 13

    arXiv 2022

  14. [14]

    Deep high-resolution representation learning for visual recognition

    Jingdong Wang, Ke Sun, Tianheng Cheng, Borui Jiang, Chaorui Deng, Yunchao Zhao, Dong Liu, Yintang Mu, Mingkui Tan, Xinggang Wang, and Wenyu Liu. Deep high-resolution representation learning for visual recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(10):3349–3364, Apr 2020. doi: 10.1109/TPAMI.2020.2983686

    work page doi:10.1109/tpami.2020.2983686 2020

  15. [15]

    A neonatal dataset and benchmark for non-contact neonatal heart rate monitoring based on spatio-temporal neural networks

    Bo Huang, Wei Chen, Chih-Lin Lin, Chia-Feng Juang, Yue Xing, Yuzhu Wang, and Jun Wang. A neonatal dataset and benchmark for non-contact neonatal heart rate monitoring based on spatio-temporal neural networks. Engineering Applications of Artificial Intelligence , 106:104447, Nov 2021. doi: 10.1016/j.engappai.2021.104447

    work page doi:10.1016/j.engappai.2021.104447 2021

  16. [16]

    Neonatal pain detection in videos using the icopevid dataset and an ensemble of descriptors extracted from gaussian of local descriptors

    Sheryl Brahnam, Loris Nanni, Steven McMurtrey, Alessandra Lumini, Ryan Brattin, Mary Slack, and Trevor Barrier . Neonatal pain detection in videos using the icopevid dataset and an ensemble of descriptors extracted from gaussian of local descriptors. Applied Computing and Informatics, 19(1/2):122–143, 2023. doi: 10.1108/ACI-05-2021-0142

    work page doi:10.1108/aci-05-2021-0142 2023

  17. [17]

    300 faces in-the-wild challenge: The first facial landmark localization challenge

    Christos Sagonas, Georgios Tzimiropoulos, Stefanos Zafeiriou, and Maja Pantic. 300 faces in-the-wild challenge: The first facial landmark localization challenge. In Proceedings of the IEEE International Conference on Computer Vision Workshops (ICCVW) , pages 397– 403, 2013

    2013

  18. [18]

    Xm2vtsdb: The extended m2vts database

    Kieron Messer, Jiri Matas, Josef Kittler, Juergen Luettin, and Gilbert Maitre. Xm2vtsdb: The extended m2vts database. In Second International Conference on Audio- and Video- Based Biometric Person Authentication (AVBPA), volume 964, pages 965–966, Mar 1999

    1999

  19. [19]

    Picallo, L

    Ralph Gross, Iain Matthews, Jeffrey Cohn, Takeo Kanade, and Simon Baker . Multi-pie. Image and Vision Computing , 28(5):807–813, 2010. ISSN 0262-8856. doi: 10.1016/j. imavis.2009.08.002. Best of Automatic Face and Gesture Recognition 2008

    work page doi:10.1016/j 2010

  20. [20]

    Jonathon Phillips, Patrick J

    P . Jonathon Phillips, Patrick J. Flynn, Todd Scruggs, Kevin W . Bowyer, Jin Chang, Kevin Hoffman, Joe Marques, Jaesik Min, and William Worek. Overview of the face recogni- tion grand challenge. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), volume 1, pages 947–954. IEEE, Jun 2005. 27

    2005

  21. [21]

    Belhumeur, David W

    Peter N. Belhumeur, David W . Jacobs, David J. Kriegman, and Neeraj Kumar . Localizing parts of faces using a consensus of exemplars. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(12):2930–2940, 2013. doi: 10.1109/TPAMI.2013.23

    work page doi:10.1109/tpami.2013.23 2013

  22. [22]

    Vu Le, Jonathan Brandt, Zhe Lin, Lubomir Bourdev, and Thomas S. Huang. Interactive facial feature localization. In European Conference on Computer Vision (ECCV) , pages 679–692, Berlin, Heidelberg, Oct 2012. Springer

    2012

  23. [23]

    Face detection, pose estimation, and landmark local- ization in the wild

    Xiaozhu Zhu and Deva Ramanan. Face detection, pose estimation, and landmark local- ization in the wild. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2879–2886. IEEE, Jun 2012

    2012

  24. [24]

    300 faces in-the-wild challenge: Database and results

    Christos Sagonas, Epameinondas Antonakos, Georgios Tzimiropoulos, Stefanos Zafeiriou, and Maja Pantic. 300 faces in-the-wild challenge: Database and results. Image and Vision Computing, 47:3–18, Mar 2016. doi: 10.1016/j.imavis.2015.11.005

    work page doi:10.1016/j.imavis.2015.11.005 2016

  25. [25]

    Burgos-Artizzu, Pietro Perona, and Piotr Dollár

    Xavier P . Burgos-Artizzu, Pietro Perona, and Piotr Dollár . Robust face landmark estimation under occlusion. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 1513–1520, 2013

    2013

  26. [26]

    The menpo facial landmark localisation challenge: A step towards the solution

    Stefanos Zafeiriou, George Trigeorgis, Grigorios Chrysos, Jiankang Deng, and Jiaxiang Shen. The menpo facial landmark localisation challenge: A step towards the solution. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Work- shops (CVPRW), pages 170–179, 2017

    2017

  27. [27]

    Look at boundary: A boundary-aware face alignment algorithm

    Wayne Wu, Chen Qian, Shuo Yang, Quan Wang, Yunde Cai, and Qiang Zhou. Look at boundary: A boundary-aware face alignment algorithm. In Proceedings of the IEEE Con- ference on Computer Vision and Pattern Recognition (CVPR), pages 2129–2138, 2018

    2018

  28. [28]

    Cobo, Frances Moultrie, Aislinn G

    Maria M. Cobo, Frances Moultrie, Aislinn G. Hauck, David Crankshaw, Victoria Monk, Caroline Hartley, Rebecca E. Fry, Sinéad Robinson, Maartje van der Vaart, Lucy Baxter, and Emma Adams. Multicentre, randomised controlled trial to investigate the effects of parental touch on relieving acute procedural pain in neonates (petal). BMJ Open, 12(7): e061841, Jul 2022

    2022

  29. [29]

    Effect of parental touch on relieving acute procedural pain in neonates and parental anx- iety (petal): a multicentre, randomised controlled trial in the uk

    Annalisa GV Hauck, Marianne van der Vaart, Eleri Adams, Luke Baxter, Aomesh Bhatt, Daniel Crankshaw, Amraj Dhami, Ria Evans Fry, Marina BO Freire, Caroline Hartley, et al. Effect of parental touch on relieving acute procedural pain in neonates and parental anx- iety (petal): a multicentre, randomised controlled trial in the uk. The Lancet Child & Adolesce...

    2024

  30. [30]

    The premature infant pain profile-revised (pipp-r): initial validation and feasibility

    Bonnie J Stevens, Sharyn Gibbins, Janet Yamada, Kimberley Dionne, Grace Lee, Céleste Johnston, and Anna Taddio. The premature infant pain profile-revised (pipp-r): initial validation and feasibility. The Clinical journal of pain , 30(3):238–243, 2014. 28

    2014

  31. [31]

    Sample and compu- tation redistribution for efficient face detection, 2021

    Jia Guo, Jiankang Deng, Alexandros Lattas, and Stefanos Zafeiriou. Sample and compu- tation redistribution for efficient face detection, 2021. URL https://arxiv.org/abs/ 2105.04714. [Preprint]. Posted 2021 May 10

    arXiv 2021

  32. [32]

    Dsfd: Dual shot face detector

    Jian Li, Yabiao Wang, Changan Wang, Ying Tai, Jun Qian, Jilin Yang, Chengjie Wang, Jinhui Li, and Feiyue Huang. Dsfd: Dual shot face detector . In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 5060–5069, 2019

    2019

  33. [33]

    Joint face detection and align- ment using multitask cascaded convolutional networks

    Kaipeng Zhang, Zhanpeng Zhang, Zhifeng Li, and Yu Qiao. Joint face detection and align- ment using multitask cascaded convolutional networks. IEEE Signal Processing Letters , 23(10):1499–1503, Oct 2016. doi: 10.1109/LSP.2016.2603342

    work page doi:10.1109/lsp.2016.2603342 2016

  34. [34]

    Computer vision annotation tool (cvat), 2023

    CVAT .ai Corporation. Computer vision annotation tool (cvat), 2023. URL https:// github.com/cvat-ai/cvat

    2023

  35. [35]

    Ultralytics yolo, Jan 2023

    Glenn Jocher, Jing Qiu, and Ayush Chaurasia. Ultralytics yolo, Jan 2023. URL https: //github.com/ultralytics/ultralytics

    2023

  36. [36]

    Pfld: A practical facial landmark detector, 2019

    Xinyao Guo, Shuling Li, Jianzhuang Yu, Jizheng Zhang, Jing Ma, Lizhuang Ma, Wei Liu, and Haibin Ling. Pfld: A practical facial landmark detector, 2019. URL https://arxiv.org/ abs/1902.10859. [Preprint]. Posted 2019 Feb 28

    Pith/arXiv arXiv 2019

  37. [37]

    Wing loss for robust facial landmark localisation with convolutional neural networks

    Zhen-Hua Feng, Josef Kittler, Muhammad Awais, Patrick Huber, and Xiao-Jun Wu. Wing loss for robust facial landmark localisation with convolutional neural networks. In Pro- ceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , pages 2235–2245, 2018

    2018

  38. [38]

    Structured landmark detection via topology- adapting deep graph learning

    Weizhe Li, Yibo Lu, Kang Zheng, Hui Liao, Chun Lin, Jing Luo, Chih-Ting Cheng, Jing Xiao, Le Lu, Chen-Feng Kuo, and Shun Miao. Structured landmark detection via topology- adapting deep graph learning. In European Conference on Computer Vision (ECCV), pages 266–283. Springer International Publishing, Cham, Aug 2020

    2020

  39. [39]

    InfantFace

    Ming Zhu, Deyu Shi, Min Zheng, and Muhammad Sadiq. Robust facial landmark detection via occlusion-adaptive deep networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 3486–3496, 2019. 29 Supplementary Material Preprint submitted to arXiv. This manuscript is a preprint and has not yet been peer reviewed...

    2019