OOIR: Observatory of International Research

Papers

(The H4-Index of International Journal of Computer Vision is 47. 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 2020-03-01 to 2024-03-01.)

Article	Citations
Knowledge Distillation: A Survey	919
The Open Images Dataset V4	768
FairMOT: On the Fairness of Detection and Re-identification in Multiple Object Tracking	570
BiSeNet V2: Bilateral Network with Guided Aggregation for Real-Time Semantic Segmentation	535
Image Matching from Handcrafted to Deep Features: A Survey	483
Deep Image Prior	338
HOTA: A Higher Order Metric for Evaluating Multi-object Tracking	315
Learning to Prompt for Vision-Language Models	311
Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation	251
Beyond Brightening Low-light Images	248
Scene Text Detection and Recognition: The Deep Learning Era	211
Image Matching Across Wide Baselines: From Paper to Practice	173
Human Action Recognition and Prediction: A Survey	160
SDNet: A Versatile Squeeze-and-Decomposition Network for Real-Time Image Fusion	158
The MVTec Anomaly Detection Dataset: A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection	145
Weakly-supervised Semantic Guided Hashing for Social Image Retrieval	135
MOTChallenge: A Benchmark for Single-Camera Multiple Target Tracking	130
Attention Guided Low-Light Image Enhancement with a Large Scale Low-Light Simulation Dataset	127
OCNet: Object Context for Semantic Segmentation	118
EfficientPS: Efficient Panoptic Segmentation	112
You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network	101
Benchmarking Low-Light Image Enhancement and Beyond	100
Deep Image Deblurring: A Survey	98
Rescaling Egocentric Vision: Collection, Pipeline and Challenges for EPIC-KITCHENS-100	96
Unsupervised Scale-Consistent Depth Learning from Video	88

Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis	82
Comparison of Full-Reference Image Quality Models for Optimization of Image Processing Systems	81
Pix2Vox++: Multi-scale Context-aware 3D Object Reconstruction from Single and Multiple Images	79
PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detection	77
Pixel-Wise Crowd Understanding via Synthetic Data	73
LaSOT: A High-quality Large-scale Single Object Tracking Benchmark	71
Unsupervised Deep Representation Learning for Real-Time Tracking	69
Unified Quality Assessment of in-the-Wild Videos with Mixed Datasets Training	69
On the Arbitrary-Oriented Object Detection: Classification Based Approaches Revisited	67
JÂA-Net: Joint Facial Action Unit Detection and Face Alignment Via Adaptive Attention	67
A Comprehensive Analysis of Weakly-Supervised Semantic Segmentation in Different Image Domains	61
Curriculum Learning: A Survey	61
Reference Pose Generation for Long-term Visual Localization via Learned Features and View Synthesis	58
Pixel-in-Pixel Net: Towards Efficient Facial Landmark Detection in the Wild	53
CLIP-Adapter: Better Vision-Language Models with Feature Adapters	52
Adaptive Channel Selection for Robust Visual Object Tracking with Discriminative Correlation Filters	52
Structure-Measure: A New Way to Evaluate Foreground Maps	49
GhostNets on Heterogeneous Devices via Cheap Operations	49
Explainability of Deep Vision-Based Autonomous Driving Systems: Review and Challenges	49
Deformable Kernel Networks for Joint Image Filtering	49
VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change	48
ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and Beyond	48
Learning Adaptive Attribute-Driven Representation for Real-Time RGB-T Tracking	47
An Exploration of Embodied Visual Exploration	47
AutoScale: Learning to Scale for Crowd Counting	47