archive
Every paper Pith has read. Search by title, abstract, or pith.
837 papers in eess.IV · page 6
-
Benchmark finds limited gains from scaling visual SSM encoders
A Controlled Benchmark of Visual State-Space Backbones with Domain-Shift and Boundary Analysis for Remote-Sensing Segmentation
-
ESFM predicts variables in data gaps while preserving physical links
Earth System Foundation Model (ESFM): A unified framework for heterogeneous data integration and forecasting
-
Schrödinger Bridge gives direct optimal paths for semantic image decoding
Optimally Bridging Semantics and Data: Generative Semantic Communication via Schr\"odinger Bridge
-
Tool-free calibration for sports cameras uses human and stick poses
Multi-Camera Self-Calibration in Sports Motion Capture: Leveraging Human and Stick Poses
-
Standard verifies provenance in medical imaging datasets
VIDS: A Verified Imaging Dataset Standard for Medical AI
-
Neural net embeds classical nonlocal denoising into one block
Learned Nonlocal Feature Matching and Filtering for RAW Image Denoising
-
Thermal imaging model detects breathing patterns at 98.8% accuracy
BreathAI: Transfer Learning-Based Thermal Imaging for Automated Breathing Pattern Recognition
-
Vision transformer ensemble reaches 96.77% AUC on deepfake test set
Towards Generalizable Deepfake Image Detection with Vision Transformers
-
Chaos map injection lifts few-shot tumor accuracy to 84.5 percent
Chaos-Enhanced Prototypical Networks for Few-Shot Medical Image Classification
-
Two-stage AI segments ten GI organs at 89 percent accuracy
A Two-Stage Deep Learning Framework for Segmentation of Ten Gastrointestinal Organs from Coronal MR Enterography
-
Learned waveforms deliver video over 2.3 kbps underwater channels
E2E-WAVE: End-to-End Learned Waveform Generation for Underwater Video Multicasting
-
Hierarchical unmixing beats standard methods on lab hyperspectral scenes
Hyperspectral Unmixing Hierarchies
-
3D-SVD matches Tucker quality for biological volumes at lower cost
Structured 3D-SVD: A Practical Framework for the Compression and Reconstruction of Biological Volumetric Images
-
Tri-stage AI tops ultrasound tasks on 27 datasets
Unified Ultrasound Intelligence Toward an End-to-End Agentic System
-
Multi-scale attention boosts 3D detection mAP by 4.78 percent
LOD-Net: Locality-Aware 3D Object Detection Using Multi-Scale Transformer Network
-
Selective squeezing on PCA modes cuts quantum resources for imaging
Resource-Efficient Quantum-Enhanced Compressive Imaging via Quantum Classical co-Design
-
Two-stage fusion predicts brain age over entire lifespan
A Two-Stage Multi-Modal MRI Framework for Lifespan Brain Age Prediction
-
Dual AI fuses CT and microscope images to classify lung cancer
Dual-Modal Lung Cancer AI: Interpretable Radiology and Microscopy with Clinical Risk Integration
-
Topology module lifts NSD scores on African brain tumor scans
Topology-Driven Fusion of nnU-Net and MedNeXt for Accurate Brain Tumor Segmentation on Sub-Saharan Africa Dataset
-
Chest CT segmentation scores drop 69% with patient-disjoint splits
CTSCAN: Evaluation Leakage in Chest CT Segmentation and a Reproducible Patient-Disjoint Benchmark
-
RelativeFlow denoises medical images using only noisy references
RelativeFlow: Taming Medical Image Denoising Learning with Noisy Reference
-
Synthetic complex MRI trains better abnormality detectors than real scans
Generative Modeling of Complex-Valued Brain MRI Data
-
Spectral SSM reaches 85.7% ImageNet accuracy without scanning
HAMSA: Scanning-Free Vision State Space Models via SpectralPulseNet
-
Smartphone camera quantifies fluorescence in 96-well plates
Design and Validation of a Low-Cost Smartphone Based Fluorescence Detection Platform Compared with Conventional Microplate Readers
-
Quality-aware AI needed for portable medical imaging to work in real settings
Portable Medical Imaging in Modern Healthcare: Fundamentals, AI-Based Taxonomy, Image Quality, and Open Challenges
-
Latent alignment lifts mental imagery decoding from fMRI
Seeing the imagined: a latent functional alignment in visual imagery decoding from fMRI data
-
Autoencoder replay lets fMRI models learn new sites without forgetting
Continual Learning for fMRI-Based Brain Disorder Diagnosis via Functional Connectivity Matrices Generative Replay
-
Frequency-domain radar tracks objects better on fast vehicles
Towards Multi-Object-Tracking with Radar on a Fast Moving Vehicle: On the Potential of Processing Radar in the Frequency Domain
-
Adaptive depth map accelerates neural implicit surface sampling
SAND: Spatially Adaptive Network Depth for Fast Sampling of Neural Implicit Surfaces
-
Learnable attention bias captures class difficulty beyond rarity
Learning Class Difficulty in Imbalanced Histopathology Segmentation via Dynamic Focal Attention
-
YOLO upgrade detects tiny drone objects 16 mAP points better
DroneScan-YOLO: Redundancy-Aware Lightweight Detection for Tiny Objects in UAV Imagery
-
AI agents generate semiconductor failure reports in under a minute
SemiFA: An Agentic Multi-Modal Framework for Autonomous Semiconductor Failure Analysis Report Generation
-
Magnitude-only beats phase for hybrid quantum SAR classification
Magnitude Is All You Need? Rethinking Phase in Quantum Encoding of Complex SAR Data
-
Phone-based setup does 3D optical tomography at 3.91 μm
Inexpensive Optical Projection Tomography on a Mobile Phone Platform
-
Uncertainty estimates in imputation raise federated X-ray AUC
Probabilistic Feature Imputation and Uncertainty-Aware Multimodal Federated Aggregation
-
Wearable ECG separates HCM from LVH at 99 percent specificity
A Wearable ECG Device for Differentiating Hypertrophic Cardiomyopathy from Acquired Left Ventricular Hypertrophy
-
Frequency bands cut attention FLOPs for million-token video diffusion
FreqFormer: Hierarchical Frequency-Domain Attention with Adaptive Spectral Routing for Long-Sequence Video Diffusion Transformers
-
Attention-enhanced DenseNet reaches 84% on three-class pneumonia X-ray task
CBAM-Enhanced DenseNet121 for Multi-Class Chest X-Ray Classification with Grad-CAM Explainability
-
License plates serve as built-in rulers for monocular car distance
Physics-Grounded Monocular Vehicle Distance Estimation Using Standardized License Plate Typography
-
License plates yield 2.3% accurate monocular vehicle distances
Physics-Grounded Monocular Vehicle Distance Estimation Using Standardized License Plate Typography
-
Neural net recovers CT attenuations near statistical bounds
Neural-Network Inversion for the Temporal CT Multi-Source Bundle Problem: Per-Bundle Statistical Limits and Near-Optimal Performance
-
Neural nets outperform classical CT estimators with patient priors
Neural-Network Inversion for the Temporal CT Multi-Source Bundle Problem: Per-Bundle Statistical Limits and Near-Optimal Performance
-
Semantic comms hits 90% accuracy with 95% data cut
Semi-Supervised Goal-Oriented Semantic Communication Framework for Foreground Classification
-
Gaze data acts as second teacher for medical segmentation
Human Gaze-based Dual Teacher Guidance Learning for Semi-Supervised Medical Image Segmentation
-
Compositional synthesis matches full supervision with 5 labels
Generative Data-engine Foundation Model for Universal Few-shot 2D Vascular Image Segmentation
-
New DSA model boosts image quality by 73 percent
VCC-DSA: A Novel Vascular Consistency Constrained DSA Imaging Model for Motion Artifact Suppression
-
Graph saliency priors from fMRI sharpen brain image reconstructions
Brain-Grasp: Graph-based Saliency Priors for Improved fMRI-based Visual Brain Decoding
-
Traffic signals localize buried fiber cables to sub-meter precision
Buried Fiber-Optic Geolocalization with Distributed Acoustic Sensing
-
Chip renders 3D Gaussian Splatting at 129 FPS in full HD
A 129FPS Full HD Real-Time Accelerator for 3D Gaussian Splatting
-
AI assigns responsibility in traffic accidents via reasoning
AITP: Traffic Accident Responsibility Allocation via Multimodal Large Language Models