Unconstrained Face Recognition: Identifying a Person of Interest From a Media Collection

By Lacey Best-Rowden, Hu Han, Charles Otto, Brendan Klare, Anil K. Jain. 2014

224 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

How Image Degradations Affect Deep CNN-Based Face Recognition? (2016, methodology): “We have applied the LFW closed-set identification protocol as our face recognition benchmark [17].”
The MegaFace Benchmark: 1 Million Faces for Recognition at Scale (2015, methodology): “Other recent studies have considered both identification as well as verification results on LFW [2, 30, 28, 27].”
Web-scale training for face identification (2014, methodology): “Fortunately, the same COTS system was recently tested [1] on a 1:N identification benchmark constructed using the images of the public Labeled Faces in the Wild (LFW) [10] dataset.”
Deeply learned face representations are sparse, selective, and robust (2014, methodology): “Therefore we further evaluate the 25 DeepID2+ nets on the closedand open-set face identification tasks on LFW, following the protocol in [3].”
DeepID3: Face Recognition with Very Deep Neural Networks (2015, unknown): “7 25 COTS-s1+s4 [1] 66 .”
Review of Deep Learning Algorithms and Architectures (2019, methodology): “Reference [64] proposes a method to utilize multiple images of the unconstrained source to enhance the recognition process.”
Targeting Ultimate Accuracy: Face Recognition via Deep Embedding (2015, methodology): “The second protocol is proposed in [3] and the protocol includes a closed-set identification task and an open-set identification task.”
Labeled Faces in the Wild: A Survey (2016, background): “[20], sometimes by augmenting LFW images with other image sets [64].”
A Light CNN for Deep Face Representation With Noisy Labels (2015, methodology): “And it also discusses more stable protocol [2] of LFW, which can indicate the robustness of face features more representatively.”
Face Spoof Detection With Image Distortion Analysis (2015, background): “However, similar to other biometric modalities [1], [2], we need to address concerns about face spoof attacks on face recognition systems, particularly in unconstrained sensing and uncooperative subject scenarios [3].”
Improved Deep Metric Learning with Multi-class N-pair Loss Objective (2016, methodology): “76 Table 3: Mean verification accuracy (VRF) with standard error, rank-1 accuracy of closed set identification and DIR@FAR=1% rate of open-set identification [1] on LFW dataset.”
Noisy Softmax: Improving the Generalization Ability of DCNN via Postponing the Early Softmax Saturation (2017, methodology): “Under unrestricted conditions, it provides 6,000 face pairs for verification protocol and, closed-set and open-set for identification protocol adopted in [2].”
Sparsifying Neural Network Connections for Face Recognition (2015, methodology): “Given the face identification protocols adopted in [3, 26, 24], we achieve 0.”
A Comprehensive Survey on Pose-Invariant Face Recognition (2015, background): “To name a few, good surveys exist for illumination-invariant face recognition [Zou et al. 2007], 3D face recognition [Bowyer et al. 2006], single image-based face recognition [Tan et al. 2006], video-based face recognition [Barr et al. 2012], and heterogeneous face recognition [Ouyang et al. 2014].”
Pushing the frontiers of unconstrained face detection and recognition: IARPA Janus Benchmark A (2015, background): “However, in practice both may be available [2].”
Bi-Directional Center-Constrained Top-Ranking for Visible Thermal Person Re-Identification (2020, background): “Related cross-modality recognition has been widely studied in VIS-NIR face recognition [12], [13], face sketch recognition [14], [15] and so on.”