IJCV 2020

159 papers
The 2020 edition of the International Journal of Computer Vision featured research focused on image restoration, deep learning architectures, and scene text analysis. A significant portion of the work addressed the enhancement of degraded visual data, including studies on Gated Fusion Networks for super-resolution and semantic-based face deblurring. The journal also explored architectural efficiency through HetConv, a method utilizing heterogeneous convolution kernels for deep convolutional neural networks. Additional research published that year examined the evolution of scene text detection and recognition, as well as techniques for constructing panoramic mosaics through global and local alignment. The edition also included a special issue dedicated to the Asian Conference on Computer Vision (ACCV) 2018 and introduced methods for testing whether multiclass classifiers operate within their intended specifications. Prolific contributors for the year included Ming-Hsuan Yang, Wei-Sheng Lai, and Fei Wang, among others. The collected works reflect a broader trend toward refining deep learning models for specialized visual tasks and improving the reliability of automated classification systems.

Top Research Directions

Research directions most represented by IJCV 2020 papers.

Generalizing to Unseen Categories via Shared Semantic Spaces 3
Resolving Geometric Symmetry Ambiguity in 3D Pose Estimation 3
Learning to Translate Images Without Paired Data 3
Differentiable Decision Trees 2
Volumetric Label Denoising via Multi-View Spatial Consensus 2
Accelerating Neural Networks via Frequency Domain Convolutions 2
Generating Synthetic Ground Truth via Simulation 2
Continuous Rotation Equivariance via Spherical Harmonics 2
Overcoming Rotation Sensitivity in 3D Deep Learning 2
Adversarial Training for Domain Invariant Representations 2
Learning Image Transformations Without Paired Data 2
Modeling the Temporal Evolution of Facial Expressions 2
Sparse Temporal Sampling for Efficient Video Understanding 2
Overcoming the 3D Supervision Bottleneck 2
Iterative Pseudo Label Refinement for Weakly Supervised Learning 2

Top Authors

AuthorPapers
Ming-Hsuan Yang7
Ran He5
Stefanos Zafeiriou4
Vittorio Ferrari4
Ling Shao3
Shuicheng YAN3
Zhenan Sun3
Adrian Hilton2
AJ Piergiovanni2
Ancong Wu2
Antonio Torralba2
Anton van den Hengel2
Baoyuan Wu2
Bernard Ghanem2
Christoph Feichtenhofer2
Christoph H. Lampert2
Dacheng Tao2
Daniel Gehrig2
Davide Scaramuzza2
Evangelos Ververas2

Impactful Papers

Papers from IJCV 2020 that shaped research clusters.

Foundational

Influential

Papers

Title
Modeling Human Motion with Quaternion-Based Neural Networks
Multimodal Image Synthesis with Conditional Implicit Maximum Likelihood Estimation
Multi-task Compositional Network for Visual Relationship Detection
Multi-task Generative Adversarial Network for Detecting Small Objects in the Wild
Pix2Vox++: Multi-scale Context-aware 3D Object Reconstruction from Single and Multiple Images
Pixelated Semantic Colorization
Predicting Intentions from Motion: The Subject-Adversarial Adaptation Approach
Product Quantization Network for Fast Visual Search
Realistic Speech-Driven Facial Animation with GANs
Real-Time Multi-person Motion Capture from Multi-view Video and IMUs
Recognizing Profile Faces by Imagining Frontal View
Rectified Wing Loss for Efficient and Robust Facial Landmark Localisation with Convolutional Neural Networks
Refractive Two-View Reconstruction for Underwater 3D Vision
Representation Learning on Unit Ball with 3D Roto-translational Equivariance
RGB-IR Person Re-identification by Cross-Modality Similarity Preservation
Robust Attentional Aggregation of Deep Feature Sets for Multi-view 3D Reconstruction
Robust Fitting in Computer Vision: Easy or Hard?
RoCGAN: Robust Conditional GAN
Scalable Person Re-Identification by Harmonious Attention
Scaling up the Randomized Gradient-Free Adversarial Attack Reveals Overestimation of Robustness Using Established Attacks
SceneFlowFields++: Multi-frame Matching, Visibility Prediction, and Robust Interpolation for Scene Flow Estimation
Scene Text Detection and Recognition: The Deep Learning Era
SeDAR: Reading Floorplans Like a Human - Using Deep Learning to Enable Human-Inspired Localisation
Semantically Coherent 4D Scene Flow of Dynamic Scenes
Semantically Tied Paired Cycle Consistency for Any-Shot Sketch-Based Image Retrieval
Semi-online Multi-people Tracking by Re-identification
Semi-supervised Semantic Mapping Through Label Propagation with Semantic Texture Meshes
Simultaneous Deep Stereo Matching and Dehazing with Feature Attention
Single Image Dehazing via Multi-scale Convolutional Neural Networks with Holistic Edges
SliderGAN: Synthesizing Expressive Face Images by Sliding 3D Blendshape Parameters
Sparse Black-Box Video Attack with Reinforcement Learning
Spatially-Adaptive Filter Units for Compact and Efficient Deep Neural Networks
Special Issue: Advances in Architectures and Theories for Computer Vision
Special Issue on ACCV 2018
Special Issue on Deep Learning for Robotic Vision
Special Issue on Generating Realistic Visual Data of Human Behavior
SSN: Learning Sparse Switchable Normalization via SparsestMax
Statistical Modeling of Craniofacial Shape and Texture
Temporal Action Detection with Structured Segment Networks
Tensorized Multi-view Subspace Representation Learning
Text2Sign: Towards Sign Language Production Using Neural Machine Translation and Generative Adversarial Networks
The Open Images Dataset V4
The Unmanned Aerial Vehicle Benchmark: Object Detection, Tracking and Baseline
Towards High Fidelity Face Frontalization in the Wild
Towards Image-to-Video Translation: A Structure-Aware Approach via Multi-stage Generative Adversarial Networks
Towards Photo-Realistic Facial Expression Manipulation
Tracking Persons-of-Interest via Unsupervised Representation Adaptation
Train Sparsely, Generate Densely: Memory-Efficient Unsupervised Training of High-Resolution Temporal GAN
Transferrable Feature and Projection Learning with Class Hierarchy for Zero-Shot Learning
Unified Binary Generative Adversarial Network for Image Retrieval and Compression

« prev Page 3/4 next »