Research map / 3D vision / Human pose estimation
HyperFace: A Deep Multi-Task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition
Rajeev Ranjan, Vishal M. Patel, Rama Chellappa
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add multitask learning (HyperFace) over single-task R-CNN_Face (ablation) | R-CNN_Face · AFW face detection | mAP (%): 90.3% -> 94% | Improved |
| add landmark-based NMS instead of linear bbox regression and NMS (ablation) | HyperFace · AFW face detection | mAP (%): +1.2% | Improved |
| add iterative region proposals (IRP) on Fast SS (ablation) | HyperFace · AFW face detection | mAP (%) | Improved |
| fuse intermediate layers (ablation) | Multitask_Face · AFLW (21 pts) | NME mean (%): 4.79 -> 4.26 | Improved |
| add multitask learning without fusion (ablation) | R-CNN_Fiducial · AFLW (21 pts) | NME mean (%): 4.76 -> 4.79 | No clear effect |
| fuse intermediate layers (ablation) | Multitask_Face · face detection AFW/PASCAL/FDDB | mAP | No clear effect |
| fuse intermediate layers (ablation) | Multitask_Face · CelebA gender | accuracy (%): 97 -> 97 | No clear effect |
| add multitask learning (ablation) | R-CNN_Gender · CelebA gender | accuracy (%): 95 -> 97 | Improved |
| add multitask learning (ablation) | R-CNN_Gender · LFWA gender | accuracy (%): 91 -> 93 | Improved |
| replace AlexNet backbone with ResNet-101 (HF-ResNet) | HyperFace · AFLW (21 pts) | NME mean (%): 4.26 -> 2.93 | Improved |
| replace AlexNet backbone with ResNet-101 (HF-ResNet) | HyperFace · IBUG 300-W (68 pts) | NME (%): 10.88 -> 8.18 | Improved |
| replace AlexNet backbone with ResNet-101 (HF-ResNet) | HyperFace · AFW face detection | mAP (%): 97.9% -> 99.4% | Improved |
Later papers that compared with this method
3 later accepted papers report HyperFace as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Hierarchical Binary CNNs for Landmark Localization with Limited Resources (TPAMI 2018 · better on 1 of 1 shared result)
- Dynamic Facial Analysis: From Bayesian Filtering to Recurrent Neural Network (CVPR 2017 · better on 6 of 6 shared results)
- Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment With Limited Resources (ICCV 2017 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multi-Task Head Pose Estimation in-the-Wild (TPAMI 2020 · similarity 0.88)
- Facial Landmark Detection with Tweaked Convolutional Neural Networks (TPAMI 2017 · similarity 0.88)
- Supervised Transformer Network for Efficient Face Detection (ECCV 2016 · similarity 0.87)
- Large Pose 3D Face Reconstruction From a Single Image via Direct Volumetric CNN Regression (ICCV 2017 · similarity 0.87)
- Robust Facial Landmark Detection via a Fully-Convolutional Local-Global Context Network (CVPR 2018 · similarity 0.87)
- AnchorFace: An Anchor-based Facial Landmark Detector Across Large Poses (AAAI 2021 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
