Surpassing Human-Level Face Verification Performance on LFW with GaussianFace

By Chaochao Lu, Xiaoou Tang. 2014

342 citations recorded by the source corpus.

Face verification remains a challenging problem in very complex conditions with large variations such as pose, illumination, expression, and occlusions. This problemis exacerbated when we rely unrealistically on a singletraining data source, which is often insufficient to coverthe intrinsically complex face variations. This paperproposes a principled multi-task learning approachbased on Discriminative Gaussian Process Latent VariableModel (DGPLVM), named GaussianFace, for faceverification. In contrast to relying unrealistically on asingle training data source, our model exploits additional data from multiple source-domains to improve the generalization performance of face verification inan unknown target-domain. Importantly, our model can adapt automatically to complex data distributions, and therefore can well capture complex face variations inherent in multiple sources. To enhance discriminative power, we introduced a more efficient equivalent form of Kernel Fisher Discriminant Analysis to DGPLVM.To speed up the process of inference and prediction, we exploited the low rank approximation method. Extensive experiments demonstrated the effectiveness of the proposed model in learning from diverse data sources and generalizing to unseen domains. Specifically, the accuracy of our algorithm achieved an impressive accuracyrate of 98.52% on the well-known and challenging Labeled Faces in the Wild (LFW) benchmark. For the first time, the human-level performance in face verification (97.53%) on LFW is surpassed.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Unsupervised Embedding Learning via Invariant and Spreading Instance Feature (2019, background): “Supervised embedding learning methods have been studied to achieve such objectives and demonstrate impressive capabilities in various vision tasks [28, 30, 53].”
FaceNet: A unified embedding for face recognition and clustering (2015, background): “Despite significant recent advances in the field of face recognition [10, 14, 15, 17], implementing face verification and recognition efficiently at scale presents serious challenges to current approaches.”
Faceless Person Recognition: Privacy Implications in Social Media (2016, background): “Like other areas of computer vision, the last years have seen a shift from classifiers based on handcrafted features and metric learning approaches [13, 14, 15, 16, 17, 18, 19] towards methods based on deep learning [20, 21, 22, 23, 24, 25, 26, 27].”
Unmasking Clever Hans predictors and assessing what machines really learn (2019, background): “They occasionally exceed human performance spectacularly in tasks as diverse as face recognition [54], traffic sign classification [55], reading subway plans [56], understanding quantum many-body systems [43,44,47,57] or playing games such as Atari 2600 games [5,6], Go [7,8], Texas hold’em poker [10”
Deep Learning Face Representation by Joint Identification-Verification (2014, methodology): “Therefore, we rely on a larger outside dataset for training, as did by all recent highperformance face verification algorithms [4, 2, 22, 21, 14].”
Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not? (2015, background): “Throughout the history of LFW benchmark, surprising improvements are obtained with recent deep learning techniques [17, 14, 13, 10, 12].”
Cognitive Science in the era of Artificial Intelligence: A roadmap for reverse-engineering the infant language-learner (2016, positive): “…go or video games: Mnih et al., 2015; Silver et al., 2016; processing speech and natural language: Amodei et al., 2015a; Ferrucci, 2012; recognizing objects and faces: He, Zhang, Ren, & Sun, 2015; Lu & Tang, 2014) and promising a revolution in manufacturing processes and human society at large.”
Deeply learned face representations are sparse, selective, and robust (2014, neutral): “26 GaussianFace [19] 98.”
Person Recognition in Personal Photo Collections (2015, background): “Both settings have seen a recent shift from sophisticated classifiers based on hand-crafted features and metric learning approaches [20, 7, 5, 29, 26, 41, 1], towards methods based on deep learning [37, 36, 43, 33, 27, 38, 21].”
DeepID3: Face Recognition with Very Deep Neural Networks (2015, methodology): “We evaluate DeepID3 networks under the LFW face verification [6] and LFW face identification [1, 18] protocols, respectively.”
Learning Pose-Aware Models for Pose-Invariant Face Recognition in the Wild (2019, background): “THERE has been a flurry of advances in face recognition in recent years, with some techniques claiming to meet [3] or even surpass [4], [5] human face verification performance.”
Secure Face Matching Using Fully Homomorphic Encryption (2018, background): “Thanks to progress in learning feature representations, automated face recognition systems are now believed to surpass human performance in some scenarios [25].”
Person Recognition in Personal Photo Collections (2017, background): “and metric learning approaches [14], [15], [16], [17].”
Effective face frontalization in unconstrained images (2014, background): “Yet in recent years recognition performance has gradually improved to the point where, once again, claims are being made for super-human face recognition capabilities (e.g., [20, 28, 29]).”
Deep Face Recognition (2015, methodology): “Shallow methods start by extracting a representation of the face image using handcrafted local image descriptors such as SIFT, LBP, HOG [5, 13, 22, 23, 32]; then they aggregate such local descriptors into an overall face descriptor by using a pooling mechanism, for example the Fisher Vector [15, 20]”
Totally Looks Like - How Humans Compare, Compared to Machines (2018, background): “This is not very surprising, as deep-learned face representations have already been reported to surpass human performance several years ago [15].”