IEEE TIP 2020

630 papers
In 2020, IEEE Transactions on Image Processing published 44 papers addressing diverse challenges in computer vision and signal processing. A portion of the research focused on hyperspectral image analysis, including material-aware vision, spectral unmixing for land cover analysis, and techniques for mitigating data scarcity. Remote sensing was a recurring theme, with studies exploring cross-view synthesis from satellite imagery and multi-scale architectures designed to manage extreme scale variance. Research directions also included advancements in visual tracking through dynamic feature weighting and deep tracklet association. The year’s edition featured work on image restoration and quality assessment, specifically evaluating restoration without ground truth and integrating chrominance into perceptual models. Efficiency was a central objective in several studies, utilizing spatiotemporal knowledge distillation for aerial video saliency and bridge distillation for low-resolution face recognition. Prolific contributors during this period included researchers such as Jia Li, Ming-Hsuan Yang, and Shengwei Zhao.

Top Research Directions

Research directions most represented by IEEE TIP 2020 papers.

Composite Loss Functions for Perceptual Image Restoration 4
Preserving Tiny Features Against Spatial Resolution Degradation 4
Cross Task Generalization in Image Restoration 3
The Limits of Hand-Crafted Priors in Image Restoration 3
Modeling Subjective and Personalized Aesthetic Preferences 3
The Perceptual Misalignment of Image Quality Metrics 3
Hierarchical Temporal Modeling for Long Video Understanding 3
Temporal Consistency Regularization for Video Processing 3
Solving Inverse Problems Without Ground Truth Training Data 3
Iterative Matrix Square Roots for Deep Learning 2
Semantic Priors for Image Restoration 2
Feature Representation Conflicts in Joint Detection and Identification 2
Variance Stabilization for Signal Dependent Noise 2
Probabilistic Boundary Refinement for Semantic Segmentation 2
Joint Latent State and Parameter Estimation 2

Top Authors

AuthorPapers
Dacheng Tao13
Ling Shao10
Jia Li9
Jiantao Zhou9
Xuelong Li9
Alan Bovik8
Meng Wang8
Wangmeng Zuo8
Bihan Wen7
Lei Zhang7
Ce Zhu6
Henry Arguello6
Huchuan Lu6
Jiaying Liu6
Jungong Han6
Shiqi Wang6
Wen Gao6
Wenhan Yang6
Wenqi Ren6
Xinbo Gao6

Impactful Papers

Papers from IEEE TIP 2020 that shaped research clusters.

Foundational

Influential

Papers

Title
Improved Robust Video Saliency Detection Based on Long-Term Spatial-Temporal Information
Improved Saliency Detection in RGB-D Images Using Two-Phase Depth Estimation and Selective Deep Fusion
Improve Person Re-Identification With Part Awareness Learning
Improving Description-Based Person Re-Identification by Multi-Granularity Image-Text Alignments
Improving Person Re-Identification With Iterative Impression Aggregation
Improving the Harmony of the Composite Image by Spatial-Separated Attention Module
Integrating Neural Networks Into the Blind Deblurring Framework to Compete With the End-to-End Learning-Based Methods
Interpretable Detail-Fidelity Attention Network for Single Image Super-Resolution
Interpret Neural Networks by Extracting Critical Subnetworks
Interweaved Prediction for Video Coding
Investigating Task-Driven Latent Feasibility for Nonconvex Image Modeling
Isosceles Constraints for Person Re-Identification
Iterative Local-Global Collaboration Learning Towards One-Shot Video Person Re-Identification
JCS: An Explainable COVID-19 Diagnosis System by Joint Classification and Segmentation
Joint Angular Refinement and Reconstruction for Single-Particle Cryo-EM
Joint Coding of Local and Global Deep Features in Videos for Visual Search
Joint Hand-Object 3D Reconstruction From a Single Image With Cross-Branch Feature Fusion
Jointly Learning Commonality and Specificity Dictionaries for Person Re-Identification
Jointly Modeling Motion and Appearance Cues for Robust RGB-T Tracking
Joint Raindrop and Haze Removal From a Single Image
Joint Stereo Video Deblurring, Scene Flow Estimation and Moving Object Segmentation
Just Noticeable Distortion-Based Perceptual Rate Control in HEVC
Just Noticeable Distortion Profile Inference: A Patch-Level Structural Visibility Learning Approach
Latent Complete Row Space Recovery for Multi-View Subspace Clustering
Latent Elastic-Net Transfer Learning
LCSCNet: Linear Compressing-Based Skip-Connecting Network for Image Super-Resolution
Learnable Descriptors for Visual Search
Learned Fast HEVC Intra Coding
Learning a Deep Dual Attention Network for Video Super-Resolution
Learning a Single Model With a Wide Range of Quality Factors for JPEG Image Artifacts Removal
Learning a Single Tucker Decomposition Network for Lossy Image Compression With Multiple Bits-per-Pixel Rates
Learning Category- and Instance-Aware Pixel Embedding for Fast Panoptic Segmentation
Learning Deeply Aggregated Alternating Minimization for General Inverse Problems
Learning Dual Encoding Model for Adaptive Visual Understanding in Visual Dialogue
Learning From Synthetic Images via Active Pseudo-Labeling
Learning Hybrid Representation by Robust Dictionary Learning in Factorized Compressed Space
Learning Interleaved Cascade of Shrinkage Fields for Joint Image Dehazing and Denoising
Learning Latent Global Network for Skeleton-Based Action Prediction
Learning Latent Low-Rank and Sparse Embedding for Robust Image Feature Extraction
Learning Layer-Skippable Inference Network
Learning Long-Term Structural Dependencies for Video Salient Object Detection
Learning Modality-Specific Representations for Visible-Infrared Person Re-Identification
Learning Nonclassical Receptive Field Modulation for Contour Detection
Learning Recurrent 3D Attention for Video-Based Person Re-Identification
Learning Rich Part Hierarchies With Progressive Attention Networks for Fine-Grained Image Recognition
Learning Sparse and Identity-Preserved Hidden Attributes for Person Re-Identification
Learning Spatial and Spatio-Temporal Pixel Aggregations for Image and Video Denoising
Learning Symmetry Consistent Deep CNNs for Face Completion
Learning to Align via Wasserstein for Person Re-Identification
Learning to Compare Relation: Semantic Alignment for Few-Shot Learning

« prev Page 7/13 next »