A survey on deep learning based face recognition

doi:10.1016/j.cviu.2019.102805

Paper

A survey on deep learning based face recognition

Published Dec 1, 2019 · G. Guo, N. Zhang

Comput. Vis. Image Underst.

Q1 SJR score

366

Citations

6

Influential Citations

Full text Semantic Scholar

Abstract

Abstract removed due to Elsevier request; this does not indicate any issues with the research. Click the full text link above to read the abstract and view the original source.

Highly Cited

Study Snapshot

Key takeawayDeep learning methods have improved face recognition by handling variations in pose, age, illumination, expression, and heterogeneous face matching.

PopulationOlder adults (50-71 years)

Sample size24

MethodsObservational

OutcomesBody Mass Index projections

ResultsSocial networks mitigate obesity in older groups.

Sign up to use Study Snapshot

Consensus is limited without an account. Create an account or sign in to get more searches and use the Study Snapshot.

Create an account

Paper

A survey on deep learning based face recognition

Sign up to use Study Snapshot

References

Citations

Increasing Interpretability of Neural Networks By Approximating Human Visual Saliency

Using human-defined regions of saliency in neural networks significantly increases interpretability and performance, but requires 80% less human annotation data than traditional methods.

Gamifying XAI: Enhancing AI Explainability for Non-technical Users through LLM-Powered Narrative Gamifications

Our prototype enhances non-technical users' AI/XAI knowledge through LLM-powered narrative gamifications, but its effectiveness in further improving understanding remains inconclusive.

Similarity learning networks uniquely identify individuals of four marine and terrestrial species

Deep learning models can effectively identify individual animals in photographs, potentially replacing or supporting manual identification efforts in conservation and management.

RVTrack: Identity Tracking by RFID-Vision Fusion with Small Training Cost

RVTrack, an RFID-vision fusion system, accurately identifies and tracks individuals with an accuracy of 90.14% using commercial RFID devices and cameras, even in complex indoor environments.

Novel Hybrid Optimization Techniques for Enhanced Generalization and Faster Convergence in Deep Learning Models: The NestYogi Approach to Facial Biometrics

NestYogi, a novel hybrid optimization algorithm, significantly improves convergence rate and overall accuracy in deep neural networks for face detection and recognition, achieving up to 98.6% accuracy on benchmark datasets.

Defending against Model Inversion Attacks via Random Erasing

Random Erasing (RE) effectively reduces Model Inversion attack accuracy while maintaining model utility, achieving state-of-the-art performance in balancing privacy and model utility across various attacks and network architectures.

A 3D U-NET BASED ON EARLY FUSION MODEL: IMPROVEMENT, COMPARATIVE ANALYSIS WITH STATE-OF-THE-ART MODELS AND FINE-TUNING

The 3D U-Net based fusion model using HSV and grayscale color spaces shows promising results in multi-organ segmentation in medical images, with the liver showing the highest Dice score.