A self-supervised method pretrains an encoder on eight PSP images per view to learn generalizable subsurface scattering representations that transfer to relighting and dense footprint reconstruction on unseen complex objects.
hub Mixed citations
U-Net: Convolutional Networks for Biomedical Image Segmentation
Mixed citation behavior. Most common role is background (43%).
99
Pith papers citing it
Background
43% of classified citations
abstract
There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net .
hub tools
citation-role summary
background 11
method 6
baseline 3
dataset 1
citation-polarity summary
claims ledger
- abstract There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segme
co-cited works
representative citing papers
A U-Net surrogate with multigroup attention pooling is trained on OpenMC sensitivity data and combined with gradient optimization to generate grid-based critical experiment geometries that achieve c_k values up to 0.97757 for HALEU fuel validation.
The work demonstrates that multi-tracer field-level SBI on galaxy and HI maps yields 2-7 times better constraints on Omega_m and sigma_8 than single-tracer or summary-statistic approaches, with 3D maps performing best.
LatentHDR generates structurally consistent panoramic HDR images by producing one scene latent with a diffusion backbone then deterministically mapping it to multiple exposure latents via a lightweight conditional head.
EchoXFlow is a new dataset of 37,125 beamspace echocardiography recordings with separable modalities, Doppler data, ECG, and clinical annotations that enables acquisition-aware learning not possible with standard scan-converted videos.
Influpaint uses generative diffusion models on image-encoded influenza data to produce realistic and diverse epidemic trajectories that match leading ensemble methods in accuracy.
VitaminP uses paired H&E-mIF data to train a model that transfers molecular boundary information, enabling accurate whole-cell segmentation directly from routine H&E histology across 34 cancer types.
A modified DCGAN with an auxiliary discriminator using the membrane factor generates stable, previously unseen funicular shells optimized for pure compression in three dimensions.
A U-Net-based ML pipeline reconstructs the complete phase field and quantized vortex charges in 2D Bose-Einstein condensates from density snapshots alone, using synthetic training data from projected Gross-Pitaevskii simulations.
Dual Triangle Attention achieves effective bidirectional attention with built-in positional inductive bias via dual triangular masks, outperforming standard bidirectional attention on position-sensitive tasks and showing strong masked language modeling results with or without positional embeddings.
Defines diffusion processes on implicit data manifolds via proximity-graph approximations to the infinitesimal generator and carré-du-champ operator, proves convergence in law to the continuous manifold process, and provides an Euler-Maruyama integrator validated on synthetic and MNIST manifolds.
A CNN-based discrete diffusion method refines sparse contours from segmentation masks using simplified denoising steps and minimal post-processing, outperforming baselines on small medical and environmental datasets while running 3.5 times faster.
A diffusion model trained on real radio galaxy images reconstructs high-fidelity interferometric observations from VLA, EHT, and ALMA simulations and outperforms CLEAN on gridded visibilities.
SemanticBridge provides a new 3D dataset for bridge component segmentation and quantifies sensor-induced domain gaps that drop model performance by up to 11.4% mIoU.
Standard visual diffusion models operating in pixel space can approximate solutions to the inscribed square, Steiner tree, and simple polygon problems.
A U-Net GAN reconstructs CMB T and E maps from Planck-like simulations with foregrounds and systematics, achieving under 1% error outside the Galactic region and demonstrating first-time correction for non-circular beams and asymmetric scans.
SinkSAM-Net uses topographic priors and SAM with coordinate-wise bounding box jittering to create pseudo-labels for iterative self-supervised training of an EfficientNetV2-UNet, reaching about 95% of fully supervised performance on sinkhole datasets.
X-Mind proposes an efficient internal visual chain-of-thought using compressed BEV sketches and recurrent block diffusion to embed predictive world models into end-to-end driving policies.
21cmEMUv3 emulates the cylindrical 21cm power spectrum via score-based diffusion and six other 21cmFAST observables via LSTM networks at sub-percent accuracy, then uses the emulator to infer a lower limit on soft-band X-ray luminosity from HERA data.
A multimodal 3D foundation model pretrained on LSM volumes via masked reconstruction and image-text alignment enables improved few-shot segmentation, classification, and deblurring.
Mask R-CNN with ResNet-50 pre-trained on MethaneAIR and fine-tuned on MethaneSAT, plus physics-informed postprocessing, yields instance-level precision 0.60/recall 0.98 at baseline, improving to 0.71/0.94 and 0.92/0.70 in two operational modes.
Normalizing flows enable all-order QED corrections in lattice scalar QED in 2-4 dimensions with reduced variance and transferability from small to large lattices.
REPA-P aligns intermediate representations in diffusion models with physical states using first-principles PDE residuals to accelerate convergence and boost out-of-distribution robustness on PDE tasks.
SegRAG is a training-free retrieval-augmented framework that extracts class-specific point prompts from a filtered DINOv3 feature bank to boost SAM3 semantic segmentation performance on standard and agricultural benchmarks.
citing papers explorer
-
From Phase to Phenomenon: Self-Supervised Learning of Subsurface Scattering with Minimal Phase-shift Inputs
A self-supervised method pretrains an encoder on eight PSP images per view to learn generalizable subsurface scattering representations that transfer to relighting and dense footprint reconstruction on unseen complex objects.
-
Inverse Critical Experiment Design via Gradient Optimization and a Multigroup Attention-Based Neural Network Architecture
A U-Net surrogate with multigroup attention pooling is trained on OpenMC sensitivity data and combined with gradient optimization to generate grid-based critical experiment geometries that achieve c_k values up to 0.97757 for HALEU fuel validation.
-
Field-level multi-tracers simulation-based inference of cosmological parameters from 3D maps
The work demonstrates that multi-tracer field-level SBI on galaxy and HI maps yields 2-7 times better constraints on Omega_m and sigma_8 than single-tracer or summary-statistic approaches, with 3D maps performing best.
-
LatentHDR: Decoupling Exposure from Diffusion via Conditional Latent-to-Latent Mapping for Text/Image-to-Panoramic HDR
LatentHDR generates structurally consistent panoramic HDR images by producing one scene latent with a diffusion backbone then deterministically mapping it to multiple exposure latents via a lightweight conditional head.
-
EchoXFlow: A Beamspace Echocardiography Dataset for Cardiac Motion, Flow, and Function
EchoXFlow is a new dataset of 37,125 beamspace echocardiography recordings with separable modalities, Doppler data, ECG, and clinical annotations that enables acquisition-aware learning not possible with standard scan-converted videos.
-
Generative diffusion models for spatiotemporal influenza forecasting
Influpaint uses generative diffusion models on image-encoded influenza data to produce realistic and diverse epidemic trajectories that match leading ensemble methods in accuracy.
-
VitaminP: cross-modal learning enables whole-cell segmentation from routine histology
VitaminP uses paired H&E-mIF data to train a model that transfers molecular boundary information, enabling accurate whole-cell segmentation directly from routine H&E histology across 34 cancer types.
-
Physics-informed, Generative Adversarial Design of Funicular Shells
A modified DCGAN with an auxiliary discriminator using the membrane factor generates stable, previously unseen funicular shells optimized for pure compression in three dimensions.
-
Machine Learning Phase Field Reconstruction in a Bose-Einstein Condensate
A U-Net-based ML pipeline reconstructs the complete phase field and quantized vortex charges in 2D Bose-Einstein condensates from density snapshots alone, using synthetic training data from projected Gross-Pitaevskii simulations.
-
Dual Triangle Attention: Effective Bidirectional Attention Without Positional Embeddings
Dual Triangle Attention achieves effective bidirectional attention with built-in positional inductive bias via dual triangular masks, outperforming standard bidirectional attention on position-sensitive tasks and showing strong masked language modeling results with or without positional embeddings.
-
Diffusion Processes on Implicit Manifolds
Defines diffusion processes on implicit data manifolds via proximity-graph approximations to the infinitesimal generator and carré-du-champ operator, proves convergence in law to the continuous manifold process, and provides an Euler-Maruyama integrator validated on synthetic and MNIST manifolds.
-
Contour Refinement using Discrete Diffusion in Low Data Regime
A CNN-based discrete diffusion method refines sparse contours from segmentation masks using simplified denoising steps and minimal post-processing, outperforming baselines on small medical and environmental datasets while running 3.5 times faster.
-
Radio-Interferometric Image Reconstruction with Denoising Diffusion Restoration Models
A diffusion model trained on real radio galaxy images reconstructs high-fidelity interferometric observations from VLA, EHT, and ALMA simulations and outperforms CLEAN on gridded visibilities.
-
SemanticBridge - A Dataset for 3D Semantic Segmentation of Bridges and Domain Gap Analysis
SemanticBridge provides a new 3D dataset for bridge component segmentation and quantifies sensor-induced domain gaps that drop model performance by up to 11.4% mIoU.
-
Visual Diffusion Models are Geometric Solvers
Standard visual diffusion models operating in pixel space can approximate solutions to the inscribed square, Steiner tree, and simple polygon problems.
-
Deep Learning for CMB Foreground Removal and Beam Deconvolution: A U-Net GAN Approach
A U-Net GAN reconstructs CMB T and E maps from Planck-like simulations with foregrounds and systematics, achieving under 1% error outside the Galactic region and demonstrating first-time correction for non-circular beams and asymmetric scans.
-
SinkSAM-Net: Knowledge-Driven Self-Supervised Sinkhole Segmentation Using Topographic Priors and Segment Anything Model
SinkSAM-Net uses topographic priors and SAM with coordinate-wise bounding box jittering to create pseudo-labels for iterative self-supervised training of an EfficientNetV2-UNet, reaching about 95% of fully supervised performance on sinkhole datasets.
-
X-Mind: Efficient Visual Chain-of-Thought via Predictive World Model for End-to-End Driving
X-Mind proposes an efficient internal visual chain-of-thought using compressed BEV sketches and recurrent block diffusion to embed predictive world models into end-to-end driving policies.
-
21cmEMUv3: a hybrid diffusion-LSTM emulator of 21cmFAST summary observables
21cmEMUv3 emulates the cylindrical 21cm power spectrum via score-based diffusion and six other 21cmFAST observables via LSTM networks at sub-percent accuracy, then uses the emulator to infer a lower limit on soft-band X-ray luminosity from HERA data.
-
A Multimodal 3D Foundation Model for Light Sheet Fluorescence Microscopy Enables Few-Shot Segmentation, Classification, and Deblurring
A multimodal 3D foundation model pretrained on LSM volumes via masked reconstruction and image-text alignment enables improved few-shot segmentation, classification, and deblurring.
-
Plume Segmentation from MethaneSAT with Cross-Sensor Transfer Learning and Physics-Informed Postprocessing
Mask R-CNN with ResNet-50 pre-trained on MethaneAIR and fine-tuned on MethaneSAT, plus physics-informed postprocessing, yields instance-level precision 0.60/recall 0.98 at baseline, improving to 0.71/0.94 and 0.92/0.70 in two operational modes.
-
Normalizing flows for all-orders QED corrections in lattice field theory
Normalizing flows enable all-order QED corrections in lattice scalar QED in 2-4 dimensions with reduced variance and transferability from small to large lattices.
-
Learning to Think in Physics: Breaking Shortcut Learning in Scientific Diffusion via Representation Alignment
REPA-P aligns intermediate representations in diffusion models with physical states using first-principles PDE residuals to accelerate convergence and boost out-of-distribution robustness on PDE tasks.
-
SegRAG: Training-Free Retrieval-Augmented Semantic Segmentation
SegRAG is a training-free retrieval-augmented framework that extracts class-specific point prompts from a filtered DINOv3 feature bank to boost SAM3 semantic segmentation performance on standard and agricultural benchmarks.
-
A General B\'ezier Tree Encoding Counterfactual Framework for Retinal-Vessel-Mediated Disease Analysis
BTECF encodes retinal vessels as Bézier trees to enable targeted, parameter-level counterfactual interventions on vessel geometry for causal analysis of vascular diseases.
-
REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
REALISTA generates semantically coherent adversarial prompts via latent-space optimization over input-dependent editing directions, achieving stronger hallucination elicitation than prior realistic attacks on open-source and reasoning LLMs.
-
EDGER: EDge-Guided with HEatmap Refinement for Generalizable Image Forgery Localization
A dual-branch system using frequency edge cues and CLIP-based synthetic patch detection for accurate, resolution-independent image forgery localization.
-
Geometry-aware Prototype Learning for Cross-domain Few-shot Medical Image Segmentation
GeoProto enriches appearance prototypes with geometric offsets from an ordinal shape branch to improve cross-domain few-shot medical image segmentation.
-
Don't Fix the Basis -- Learn It: Spectral Representation with Adaptive Basis Learning for PDEs
ABLE learns a spatially adaptive Parseval frame from data via an ancillary density to replace fixed bases in spectral neural operators for PDEs.
-
TRAS: An Interactive Software for Tracing Tree Ring Cross Sections
TRAS is an interactive GUI tool integrating CS-TRD, DeepCS-TRD, and INBD for automatic tree ring detection and measurement in cross-sectional images, evaluated on 18 Pinus taeda samples with 81% F-score and 20% manual effort reduction.
-
Diffusion model for SU(N) gauge theories
Implicit score matching trains diffusion models that successfully sample SU(3) Wilson gauge configurations on lattices, with a Hamiltonian-dynamics corrector needed for strong coupling.
-
Leveraging Image Generators to Address Training Data Scarcity: The Gen4Regen Dataset for Forest Regeneration Mapping
Mixing real UAV imagery with 2101 AI-generated image-mask pairs improves semantic segmentation F1 scores for fine-grained forest species by over 15 percentage points overall and up to 30 points for rare classes.
-
A CNN--Transformer Denoiser for low-$S/N$ Galaxy Spectra: Stellar Population Recovery in Synthetic Tests
A hybrid CNN-Transformer denoiser trained on synthetic spectra substantially reduces noise and improves stellar population recovery for low-S/N galaxy observations in controlled tests.
-
Approaching human parity in the quality of automated organoid image segmentation
A composite SAM-based method segments organoid images with accuracy matching or approaching inter-observer variability among human annotators.
-
When Less Is More: Simplicity Beats Complexity for Physics-Constrained InSAR Phase Unwrapping
A vanilla U-Net with 7.76M parameters achieves R²=0.834 and RMSE=1.01 cm on a global InSAR benchmark, beating larger attention models by 34% in R² and 51% in RMSE while running 2.5× faster.
-
MG-NECOLA: A Field-Level Emulator for $f(R)$ Gravity and Massive Neutrino Cosmologies
A field-level CNN emulator converts MG-PICOLA runs into near N-body accuracy for f(R) gravity and neutrino cosmologies, achieving sub-percent errors on power spectra and bispectra while generalizing beyond its training set.
-
From Boundaries to Semantics: Prompt-Guided Multi-Task Learning for Petrographic Thin-section Segmentation
Petro-SAM adapts SAM via a Merge Block for polarized views plus multi-scale fusion and color-entropy priors to jointly achieve grain-edge and lithology segmentation in petrographic images.
-
Self-supervised Pretraining of Cell Segmentation Models
DINOCell achieves a SEG score of 0.784 on LIVECell by self-supervised domain adaptation of DINOv2, improving 10.42% over SAM-based models and showing strong zero-shot transfer.
-
GIF: A Conditional Multimodal Generative Framework for IR Drop Imaging in Chip Layouts
GIF fuses geometrical image features and logical graph topology in a conditional diffusion model to generate high-quality IR drop images for chip layouts, outperforming prior ML methods on CircuitNet-N28 with SSIM 0.78, Pearson 0.95, PSNR 21.77, and NMAE 0.026.
-
ELT: Elastic Looped Transformers for Visual Generation
Elastic Looped Transformers share weights across recurrent blocks and apply intra-loop self-distillation to deliver 4x parameter reduction while matching competitive FID and FVD scores on ImageNet and UCF-101.
-
MRI-to-CT synthesis using drifting models
Drifting models outperform diffusion, CNN, VAE, and GAN baselines in MRI-to-CT synthesis on two pelvis datasets with higher SSIM/PSNR, lower RMSE, and millisecond one-step inference.
-
SHANG++: Robust Stochastic Acceleration under Multiplicative Noise
SHANG++ delivers faster convergence and stronger robustness to multiplicative noise in stochastic optimization for both convex and strongly convex problems, with explicit parameters and competitive deep-learning results.
-
Forecasting implied volatility surface with generative diffusion models
A conditioned diffusion model with SNR-weighted arbitrage penalty generates one-day-ahead arbitrage-free implied volatility surfaces and outperforms baselines on market data.
-
Recovering Sub-threshold S-wave Arrivals in Deep Learning Phase Pickers via Shape-Aware Loss
A shape-aware loss strategy recovers sub-threshold S-wave arrivals in deep learning seismic phase pickers by treating labels as coherent shapes, achieving a 64% increase in effective detections.
-
DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions
DAWM introduces a modular diffusion world model with an inverse dynamics model to produce complete synthetic transitions that improve conservative offline RL algorithms like TD3BC and IQL on D4RL tasks.
-
Flow marching for a generative PDE foundation model
Flow Marching jointly samples noise and physical time to learn a velocity field for generative PDE modeling, paired with a latent autoencoder and efficient transformer for large-scale pretraining on 2.5M trajectories.
-
Label Dropout: Improved Deep Learning Echocardiography Segmentation Using Multiple Datasets With Domain Shift and Partial Labelling
Label dropout mitigates shortcut learning in multi-dataset partially labelled echocardiography segmentation, improving Dice scores by 62% and 25% on two cardiac structures.
-
SDXL-Lightning: Progressive Adversarial Diffusion Distillation
SDXL-Lightning uses progressive adversarial distillation to reach new state-of-the-art quality in one-step and few-step 1024px text-to-image generation from the SDXL base model.
-
Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation
A low-cost whole-body teleoperation system enables effective imitation learning for complex bimanual mobile manipulation by co-training on mobile and static demonstration datasets.
-
Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets
Stable Video Diffusion scales latent video diffusion models via text-to-image pretraining, video pretraining on curated data, and high-quality finetuning to produce competitive text-to-video and image-to-video results while enabling motion LoRA and multi-view 3D applications.