Reconstruction-based IMVC is structurally untrainable when complete-sample proportion falls near zero; CRAFT escapes that bound via per-sample attention-masked fusion trained once on complete data.
hub Mixed citations
IEEE Transactions on Image Processing 26(5), 2274–2285 (2017)
Mixed citation behavior. Most common role is background (50%).
27
Pith papers citing it
Background
50% of classified citations
hub tools
citation-role summary
background 3
method 2
dataset 1
citation-polarity summary
representative citing papers
A mixture model with adaptive KDE and per-image cross-validation raises estimated human fixation consistency by 5-15% median log-likelihood and up to 2 AUC points over fixed-bandwidth Gaussian baselines.
PAMELA provides a multi-user rating dataset and personalized reward model that predicts individual image preferences more accurately than prior population-level aesthetic models.
LucidFlux is a caption-free image restoration method that conditions a Flux.1 diffusion transformer with a dual-branch module from the degraded input and a proxy restoration plus SigLIP semantic features to outperform baselines on synthetic and real-world data.
A Debye CZT-based wave-optics pipeline generates lens-diverse synthetic defocus blur datasets that improve cross-device deblurring generalization over existing real and synthetic data.
Introduces the TUB dataset of 1320 real turbid underwater images and PCD metric showing strong correlation with instance segmentation performance where standard metrics fail.
SENTRY is a plug-and-play module that replaces confidence-based memory writes with neighbor-aware cycle-consistent validation in SAM2 trackers, yielding new zero-shot SOTA results on LaSOT, GOT-10k and other benchmarks.
A teacher-student reward model learns reasoning-conditioned score distributions for text-to-image images, yielding ~89% preference accuracy and a 41% net human-preference gain when used for generator optimization.
CoVEBench is a new benchmark showing that existing text-guided video editing models frequently fail on compositional instructions involving simultaneous subject, action, and camera changes.
UHD-GCN-BIQA models structural dependencies among sampled patches via a hybrid kNN graph and residual graph convolutions to achieve competitive PLCC and SRCC with the lowest RMSE on the UHD-IQA benchmark for blind ultra-high-definition image quality assessment.
Deep UCSL uses a contrastive EM loss on patient-control labels to isolate disease-driven subgroups in medical imaging by suppressing shared healthy variability.
MG-IQA trains vision-language models with attribute-aware RL2R and a multi-dimensional Thurstone reward model to jointly predict overall quality and fine-grained attributes, reporting 2.1% average SRCC gains on eight IQA benchmarks.
RealLiFe optimizes multi-plane images with HSGD to deliver real-time light field reconstruction from sparse views, claiming 100x speedup over offline methods and 2 dB PSNR gain over online ones.
Fewer than d/2 errors in line sums can be corrected in discrete tomography, with the bound shown to be optimal.
SpecTrack allocates variable-capacity mixture-of-experts to multispectral search regions using spectral prompt routing, achieving competitive AUC on three MSI/HSI benchmarks and GOT-10k.
End-to-end pipeline uses ResViT-2.5D to synthesize post-resection MRI from ioUS then anchors deformable registration, yielding 5.86 mm TRE on 14 ReMIND subjects while producing an integrated whole-brain volume reflecting intraoperative state.
Authors create psychometrically scaled image sets from human tests on denoised photos and provide a HaarPSI threshold for choosing denoising parameters based on perceived similarity.
DSCC groups spectrally similar and spatially close pixels into supertokens using multi-criteria distance and soft labels, then classifies at the token level to achieve 0.728 CF1 at 197.75 FPS on WHU-OHS.
RoomRecon delivers a real-time mobile system for high-quality textured 3D room reconstructions that combines AR-guided imaging with generative AI texturing focused on permanent structures and claims to outperform prior methods in quality and speed.
DAT combines a small-large model cascade with fine-tuning and bandwidth-aware multi-stream transmission to deliver high-accuracy event recognition and low-latency alerts for video streams in edge-cloud systems.
A pointwise multivariate information-driven sampling method generates reduced datasets that preserve statistical associations among variables for effective feature queries and analysis.
The paper proposes the Aesthetic Multi-Attribute Network (AMAN) that jointly predicts captions and scores for five aesthetic attributes using a new weakly-labeled dataset created via knowledge transfer.
A DenseNet201 base model trained on a constructed plant leaf disease dataset outperforms baselines and enables faster, more robust transfer learning with less data than general models.
A literature survey on abstract concept recognition in videos that catalogs prior tasks and datasets while advocating for foundation models and reuse of decades of community experience.
citing papers explorer
-
Rethinking Incompleteness: Formalizing Protocol Divergence and Train-Once Learning for Robust IMVC
Reconstruction-based IMVC is structurally untrainable when complete-sample proportion falls near zero; CRAFT escapes that bound via per-sample attention-masked fusion trained once on complete data.
-
Raising the Ceiling: Better Empirical Fixation Densities for Saliency Benchmarking
A mixture model with adaptive KDE and per-image cross-validation raises estimated human fixation consistency by 5-15% median log-likelihood and up to 2 AUC points over fixed-bandwidth Gaussian baselines.
-
Personalizing Text-to-Image Generation to Individual Taste
PAMELA provides a multi-user rating dataset and personalized reward model that predicts individual image preferences more accurately than prior population-level aesthetic models.
-
LucidFlux: Caption-Free Photo-Realistic Image Restoration via a Large-Scale Diffusion Transformer
LucidFlux is a caption-free image restoration method that conditions a Flux.1 diffusion transformer with a dual-branch module from the degraded input and a proxy restoration plus SigLIP semantic features to outperform baselines on synthetic and real-world data.
-
Realistic Compound-Lens Defocus Blur Synthesis
A Debye CZT-based wave-optics pipeline generates lens-diverse synthetic defocus blur datasets that improve cross-device deblurring generalization over existing real and synthetic data.
-
Beyond Aesthetics: Quantifying Information Loss in Turbid Scenes
Introduces the TUB dataset of 1320 real turbid underwater images and PCD metric showing strong correlation with instance segmentation performance where standard metrics fail.
-
SENTRY: SAM2-Enhanced Neighbor-Aware and Temporally Reasoned Memory for Visual Tracking
SENTRY is a plug-and-play module that replaces confidence-based memory writes with neighbor-aware cycle-consistent validation in SAM2 trackers, yielding new zero-shot SOTA results on LaSOT, GOT-10k and other benchmarks.
-
Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions
A teacher-student reward model learns reasoning-conditioned score distributions for text-to-image images, yielding ~89% preference accuracy and a 41% net human-preference gain when used for generator optimization.
-
CoVEBench: Can Video Editing Models Handle Complex Instructions?
CoVEBench is a new benchmark showing that existing text-guided video editing models frequently fail on compositional instructions involving simultaneous subject, action, and camera changes.
-
Ultra-High-Definition Image Quality Assessment via Graph Representation Learning
UHD-GCN-BIQA models structural dependencies among sampled patches via a hybrid kNN graph and residual graph convolutions to achieve competitive PLCC and SRCC with the lowest RMSE on the UHD-IQA benchmark for blind ultra-high-definition image quality assessment.
-
Automatic Discovery of Disease Subgroups by Contrasting with Healthy Controls
Deep UCSL uses a contrastive EM loss on patient-control labels to isolate disease-driven subgroups in medical imaging by suppressing shared healthy variability.
-
Multi-Granularity Reasoning for Image Quality Assessment via Attribute-Aware Reinforcement Learning to Rank
MG-IQA trains vision-language models with attribute-aware RL2R and a multi-dimensional Thurstone reward model to jointly predict overall quality and fine-grained attributes, reporting 2.1% average SRCC gains on eight IQA benchmarks.
-
RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent
RealLiFe optimizes multi-plane images with HSGD to deliver real-time light field reconstruction from sparse views, claiming 100x speedup over offline methods and 2 dB PSNR gain over online ones.
-
Error Correction for Discrete Tomography
Fewer than d/2 errors in line sums can be corrected in discrete tomography, with the bound shown to be optimal.
-
SpecTrack: Spectral Prompt Guided Adaptive Experts for Multispectral Object Tracking
SpecTrack allocates variable-capacity mixture-of-experts to multispectral search regions using spectral prompt routing, achieving competitive AUC on three MSI/HSI benchmarks and GOT-10k.
-
What neurosurgeons need to see: synthetic intra-operative MRI from ultrasound for brain-shift compensation in brain tumour surgery
End-to-end pipeline uses ResViT-2.5D to synthesize post-resection MRI from ioUS then anchors deformable registration, yielding 5.86 mm TRE on 14 ReMIND subjects while producing an integrated whole-brain volume reflecting intraoperative state.
-
Mathematical framework for perception-driven parameter choice in image denoising
Authors create psychometrically scaled image sets from human tests on denoised photos and provide a HaarPSI threshold for choosing denoising parameters based on perceived similarity.
-
Hyperspectral Image Classification via Efficient Global Spectral Supertoken Clustering
DSCC groups spectrally similar and spatially close pixels into supertokens using multi-criteria distance and soft labels, then classifies at the token level to achieve 0.728 CF1 at 197.75 FPS on WHU-OHS.
-
RoomRecon: High-Quality Textured Room Layout Reconstruction on Mobile Devices
RoomRecon delivers a real-time mobile system for high-quality textured 3D room reconstructions that combines AR-guided imaging with generative AI texturing focused on permanent structures and claims to outperform prior methods in quality and speed.
-
DAT: Dual-Aware Adaptive Transmission for Efficient Multimodal LLM Inference in Edge-Cloud Systems
DAT combines a small-large model cascade with fine-tuning and bandwidth-aware multi-stream transmission to deliver high-accuracy event recognition and low-latency alerts for video streams in edge-cloud systems.
-
Multivariate Pointwise Information-Driven Data Sampling and Visualization
A pointwise multivariate information-driven sampling method generates reduced datasets that preserve statistical associations among variables for effective feature queries and analysis.
-
Aesthetic Attributes Assessment of Images
The paper proposes the Aesthetic Multi-Attribute Network (AMAN) that jointly predicts captions and scores for five aesthetic attributes using a new weakly-labeled dataset created via knowledge transfer.
-
Developing a Strong Pre-Trained Base Model for Plant Leaf Disease Classification
A DenseNet201 base model trained on a constructed plant leaf disease dataset outperforms baselines and enables faster, more robust transfer learning with less data than general models.
-
Looking Beyond the Obvious: A Survey on Abstract Concept Recognition for Video Understanding
A literature survey on abstract concept recognition in videos that catalogs prior tasks and datasets while advocating for foundation models and reuse of decades of community experience.
-
Monitoring road infrastructures from satellite images in Greater Maputo
Object-oriented RGB pixel distribution analysis from satellite images classifies paved versus unpaved roads in Greater Maputo.
-
RGB-D image-based Object Detection: from Traditional Methods to Deep Learning Techniques
A survey of RGB-D object detection from traditional hand-crafted features with machine learning to deep learning techniques.
- Classical Versus Deep Mirror-Symmetry Scoring: A Benchmark of Thirteen Methods