Computer Vision and Image Understanding

Papers
(The H4-Index of Computer Vision and Image Understanding is 27. The table below lists those papers that are above that threshold based on CrossRef citation counts [max. 250 papers]. The publications cover those that have been published in the past four years, i.e., from 2020-05-01 to 2024-05-01.)
ArticleCitations
Skeleton-based action recognition via spatial and temporal transformer networks143
Deep 3D human pose estimation: A review138
Video anomaly detection and localization via Gaussian Mixture Fully Convolutional Variational Autoencoder126
Pyramid Channel-based Feature Attention Network for image dehazing109
Deep learning for deepfakes creation and detection: A survey96
Pros and cons of GAN evaluation measures: New developments88
A review of 3D human pose estimation algorithms for markerless motion capture79
Fake face detection via adaptive manipulation traces extraction network66
TCLR: Temporal contrastive learning for video representation66
Infrared and visible image fusion via gradientlet filter50
A comprehensive review of past and present image inpainting methods48
Single-image deblurring with neural networks: A comparative survey47
CUFD: An encoder–decoder network for visible and infrared image fusion based on common and unique feature decomposition46
Knowledge distillation for incremental learning in semantic segmentation45
High-level prior-based loss functions for medical image segmentation: A survey45
Age estimation from faces using deep learning: A comparative analysis37
Visual complexity analysis using deep intermediate-layer features36
Nighttime image dehazing based on Retinex and dark channel prior using Taylor series expansion36
Human action recognition in drone videos using a few aerial training examples33
Multi-focus image fusion approach based on CNP systems in NSCT domain33
Visual object tracking: A survey31
Learning deep edge prior for image denoising30
Detection of Face Recognition Adversarial Attacks29
Adversarial examples for replay attacks against CNN-based face recognition with anti-spoofing capability29
A survey on bias in visual datasets28
Curriculum self-paced learning for cross-domain object detection28
End-to-end deep learning-based fringe projection framework for 3D profiling of objects27
0.044042110443115