Unconstrained Face Detection and Open-Set Face Recognition Challenge

  title={Unconstrained Face Detection and Open-Set Face Recognition Challenge},
  author={Manuel G{\"u}nther and Peiyun Hu and Christian Herrmann and Chi-Ho Chan and Min Jiang and Shufan Yang and Akshay Raj Dhamija and Deva Ramanan and J{\"u}rgen Beyerer and Josef Kittler and Mohamad Al Jazaery and Mohammad Iqbal Nouyed and Guodong Guo and Cezary Stankiewicz and Terrance E. Boult},
  journal={2017 IEEE International Joint Conference on Biometrics (IJCB)},
Face detection and recognition benchmarks have shifted toward more difficult environments. [] Key Result By contrast, open-set face recognition is currently weak and requires much more attention.

Figures and Tables from this paper

A Comparative Analysis of the Face Recognition Methods in Video Surveillance Scenarios

This study presents comparative benchmark tables for the state-of-art face recognition methods by testing them with same backbone architecture in order to focus only on the face recognition solution instead of network architecture.

Face Detection on Surveillance Images

Experimental results show that, the proposed method that leverages body information to detect faces, is the most realistic approach in terms of accuracy, false alarms and average detection time, when surveillance scenario is in consideration.

Context-Aware Face Detection for Occluded Faces

This work proposed in this work an occluded faces dataset in high degree of occlusion along with context-based labelling, which is a useful technique that let CNN to learn more features during training.

Surveillance Face Recognition Challenge

It is shown that the current state of the arts are still far from being satisfactory to tackle the under-investigated surveillance FR problem in practical forensic scenarios, and introduces a new Surveillance Face Recognition Challenge, which is the largest and more importantly the only true surveillance FR benchmark, where low-resolution images are not synthesised by artificial down-sampling of native high- resolution images.

Unconstrained Face Identification using Ensembles trained on Clustered Data

A scalable open-set face identification approach to galleries composed of hundreds and thousands of subjects composed of clustering and ensemble of binary learning algorithms that estimates when query face samples belong to the face gallery and then retrieves their correct identity.

Pushing the Limits of Unconstrained Face Detection: a Challenge Dataset and Baseline Results

This work identifies the next set of challenges that requires attention from the research community and collects a new dataset of face images that involve these issues such as weather-based degradations, motion blur, focus blur and several others.

Towards a Reliable Face Recognition System

There is a significant reliance by these methods on preprocessing for optimum performance of face detection and recognition models, particularly HOG, YOLO and MTCNN.

Occluded Face Detection, Face in Niqab Dataset

A dataset of occluded faces named faces in niqab is proposed and experimental results indicated the poor performance of state-of-the-art face detection algorithms when tested in this dataset.

YOLO-FD: YOLO for Face Detection

This paper proposes a customized version of the state-of-the-art object detector, YOLOv3, for face detection, aimed at building a real-time, accurate model capable of detecting faces as small as 16 pixels in 34 FPS.



WIDER FACE: A Face Detection Benchmark

There is a gap between current face detection performance and the real world requirements, and the WIDER FACE dataset, which is 10 times larger than existing datasets is introduced, which contains rich annotations, including occlusions, poses, event categories, and face bounding boxes.

Toward Open-Set Face Recognition

This paper formulates an open-set face identification protocol, based on the canonical labeled faces in the wild (LFW) dataset, and introduces the concepts of known unknowns and unknown unknowns to the biometric community.

Large scale unconstrained open set face database

  • Archana SapkotaT. Boult
  • Computer Science
    2013 IEEE Sixth International Conference on Biometrics: Theory, Applications and Systems (BTAS)
  • 2013
This paper addresses large scale, unconstrained, open set face recognition, which exhibits the properties of operational face recognition scenarios and releases part 1 of the database, which consists of 6,337 images from 308 subjects.

SCface – surveillance cameras face database

A database of static images of human faces taken in uncontrolled indoor environment using five video surveillance cameras of various qualities to enable robust face recognition algorithms testing, emphasizing different law enforcement and surveillance use case scenarios is described.

Pushing the frontiers of unconstrained face detection and recognition: IARPA Janus Benchmark A

Baseline accuracies for both face detection and face recognition from commercial and open source algorithms demonstrate the challenge offered by this new unconstrained benchmark.

The 2013 face recognition evaluation in mobile environment

An evaluation on face recognition in mobile environment using two verification metrics to assess the usability of the algorithms in mobile devices with limited resources.

Low-resolution Convolutional Neural Networks for video face recognition

A face recognition method based on a Convolutional Neural Network with a manifold-based track comparison strategy for low-resolution video face recognition that outperforms local image descriptors as well as the state-of-the-art VGG-Face network in this domain.

The challenge of face recognition from digital point-and-shoot cameras

The Point-and-Shoot Face Recognition Challenge (PaSC) is introduced, featuring 9,376 still images of 293 people balanced with respect to distance to the camera, alternative sensors, frontal versus not-frontal views, and varying location.

Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks

A deep cascaded multitask framework that exploits the inherent correlation between detection and alignment to boost up their performance and achieves superior accuracy over the state-of-the-art techniques on the challenging face detection dataset and benchmark.

Low-Quality Video Face Recognition with Deep Networks and Polygonal Chain Distance

A face recognition system addressing the typical issues such as motion blur, noise or compression artifacts to improve low-quality recognition rates is suggested and leads to a performance improvement on in-the-wild surveillance data compared to conventional local feature approaches or the state-of- the-art high-resolution VGG-Face network.