Quaternion CNN outperforms traditional CNN in color image reconstruction.
problem Efficient image processing with small or heterogeneous datasets.
method Introducing quaternion-valued convolutional neural networks (QCNN) to learn internal and external relations.
result QCAE outperforms CAE in reconstructing unseen color images.
This paper explores neural networks for colorizing grayscale images.
problem Colorizing grayscale images using neural networks.
method Comparison of existing and novel generative models (CVAE, CWGAN-GP, AGE, IVAE) trained on CIFAR-10 images.
result CVAE with L1 reconstruction loss and IVAE achieve the highest Inception Score (IS).
Recent progress on many imaging and vision tasks has been driven by the use of deep feed-forward neural networks, which are trained by propagating gradients of a loss defined on the final output, back through the network up to the first layer that operates directly on the image. We propose back-propagating one step fur…
V1 cortex reconstructs images as Poisson equation solutions with varying weights.
problem Reconstructing images from V1 cortical cell receptive profiles.
method Solves a heterogeneous Poisson equation with varying weights representing neural connectivity.
result Reconstructions converge to homogeneous solutions using homogenization techniques.
Faster and accurate JPEG2000 image classification without reconstruction.
problem Efficiently classify j2k-compressed images without reconstructing them.
method Train a deep CNN using DWT coefficients directly from j2k-compressed images, using different augmentation techniques.
result Achieved faster and more accurate classification of j2k images without additional computation.
Improved image recovery with minimal data using untrained neural networks.
problem Solving inverse problems with limited data.
method Pre-training neural networks with a small number of examples to improve performance.
result Performance increases as data increases, matching generative models with less than 1% of training data.
DEMUD-VIS detects novel image content and explains it visually.
problem Detecting and explaining novel image content in large datasets.
method Uses CNN for feature extraction, reconstruction error for novelty detection, and up-convolutional networks for image reconstruction.
result Demonstrates visual explanations of novel image content on diverse datasets.
The ability to characterize the color content of natural imagery is an important application of image processing. The pixel by pixel coloring of images may be viewed naturally as points in color space, and the inherent structure and distribution of these points affords a quantization, through clustering, of the color i…
Develops new methods to create imperceptible image changes that fool classifiers.
problem Improving the robustness of image classifiers by creating subtle changes undetectable to humans.
method Two methods: Edge-Aware and Color-Aware, designed to reduce detectability of image perturbations.
result Demonstrated that the new methods effectively cause misclassification and are computationally efficient.
This work reconstructs knot invariants from Alexander polynomials, proving consistency with known theorems.
problem Reconstructing knot invariants from Alexander polynomials.
method Quantization, deformation, and rewriting of Alexander polynomials.
result Derives new formulae for colored superpolynomials and proves consistency with Melvin-Morton-Rozansky theorem.
Paper proposes an algorithm to reconstruct optimal model structure from graph adjacency matrix.
problem Optimal model structure reconstruction from weighted colored graph adjacency matrix.
method Uses prize-collecting Steiner tree algorithm to reconstruct minimum spanning tree.
result Demonstrates the effectiveness of the prize-collecting Steiner tree algorithm for model structure reconstruction.
A new method classifies color images using quaternion algebra.
problem Classifying color images with preserved intrinsic relationships.
method LSQMM model with quaternion nuclear norm regularization and ADMM algorithm.
result LSQMM outperforms state-of-the-art methods in classification accuracy and efficiency.
New algorithm predicts missing matrix entries using side information, outperforming existing methods.
problem Learning a partially observed matrix with side information.
method Mixed-projection ADMM algorithm for optimization.
result Our algorithm achieves 2.3% lower objective value and 41% lower reconstruction error than benchmarks.
A novel approach for augmenting histopathological images by blending Gaussian-Laplacian pyramids.
problem Data imbalance and inter-patient variability in histopathological images.
method Image blending using Gaussian-Laplacian pyramids to distribute inter-patient variability.
result Promising gains in performance compared to existing data augmentation techniques.
Generative modeling over natural images is one of the most fundamental machine learning problems. However, few modern generative models, including Wasserstein Generative Adversarial Nets (WGANs), are studied on manifold-valued images that are frequently encountered in real-world applications. To fill the gap, this pape…
New results on inferring hidden states in trackable weak models.
problem Inferring hidden states in trackable weak models.
method Analyzing strongly-connected trackable weak models and reconstructing branch choices.
result The number of hypotheses in strongly-connected trackable models is bounded by a constant.
MetalGAN uses meta-learning and clustering to colorize images with little data.
problem Colorizing images with limited data.
method Adversarial training and meta-learning with cluster-based dataset division.
result The method achieves high-quality colorization with minimal data.
Paper tackles depth estimation and optic disc-cup segmentation from color fundus images.
problem Depth estimation and optic disc-cup segmentation from color fundus images.
method Uses fully convolutional networks for monocular retinal depth estimation and optic disc-cup segmentation.
result Demonstrates improved accuracy in depth estimation and optic disc-cup segmentation.
Automates hair color digitization using imaging and deep learning.
problem Challenges in capturing and rendering realistic hair colors.
method Combines imaging, path-tracing, and self-supervised machine learning.
result Accurately captures and renders hair color with synthetic images.
Variational auto-encoders (VAEs) are a popular and powerful deep generative model. Previous works on VAEs have assumed a factorized likelihood model, whereby the output uncertainty of each pixel is assumed to be independent. This approximation is clearly limited as demonstrated by observing a residual image from a VAE …
Spectral images captured by satellites and radio-telescopes are analyzed to obtain information about geological compositions distributions, distant asters as well as undersea terrain. Spectral images usually contain tens to hundreds of continuous narrow spectral bands and are widely used in various fields. But the vast…
IAGAN method improves medical image reconstruction by incorporating adaptive GAN priors.
problem Reconstructing high-fidelity medical images from incomplete data.
method Image-adaptive GAN-based reconstruction method (IAGAN).
result IAGAN can recover fine structures relevant for medical diagnosis.
CoRAS adapts image acquisition rates for accurate reconstruction.
problem Determining when enough measurements are collected for accurate image reconstruction.
method Adaptive acquisition rate selection based on reconstruction error probability.
result CoRAS achieves target stopping-time coverage with fewer measurements.
Paper tackles deep learning's data needs with rule-based augmentation for cartoon coloring.
problem Deep learning's need for large labeled datasets.
method Rule-based augmentation for small datasets, applied to image translation.
result Automated cartoon coloring with limited data achieved.
OTRE uses OT to improve retinal images, outperforming existing methods.
problem Improving quality of non-mydriatic retinal images for accurate diagnoses.
method OT theory for image-to-image translation, regularization by enhancing.
result OTRE outperforms state-of-the-art methods on various retinal image tasks.
Deep learning improves demosaicing but edge devices struggle.
problem Edge devices struggle with deep learning-based demosaicing.
method Exhaustive search of deep neural network architectures to find the best balance between performance and model complexity.
result Found architectures that outperform state-of-the-art demosaicing models on edge devices.
Two-layer model sparsifies image residuals for CT image reconstruction.
problem Image reconstruction from limited and corrupted data.
method Pre-learning a two-layer sparsifying transform model with block coordinate descent optimization.
result Preliminary experiments show the two-layer model improves CT image reconstruction from low-dose measurements.
Capsule models detect adversarial images by reconstructing from top-level capsules.
problem Detecting adversarial images that look like a typical member of the predicted class.
method Capsule models trained to reconstruct images from pose parameters and identity of the correct top-level capsule.
result Setting a threshold on reconstruction error effectively detects adversarial images.
A hybrid network improves MRI reconstruction from undersampled data.
problem Reducing MRI acquisition time while maintaining image quality.
method Hybrid architecture combining k-space and image domains using complex-valued and real-valued U-nets.
result Hybrid approach outperformed image-only deep learning methods in hard-to-reconstruct regions.
New method reconstructs images from fMRI data using unlabeled data.
problem Challenges in acquiring labeled data for fMRI-to-image reconstruction.
method Self-supervised training with Encoder-Decoder and Decoder-Encoder networks.
result Reconstruction network adapts to new unlabeled test data.
Vision impairment due to pathological damage of the retina can largely be prevented through periodic screening using fundus color imaging. However the challenge with large scale screening is the inability to exhaustively detect fine blood vessels crucial to disease diagnosis. In this work we present a computational ima…
DEER network improves few-view breast CT image reconstruction efficiency and quality.
problem Efficient and high-quality few-view breast CT image reconstruction.
method Deep Efficient End-to-end Reconstruction (DEER) network with low model complexity.
result DEER network achieves competitive image quality with significantly fewer parameters compared to state-of-the-art methods.
PC-RNN reconstructs MRI images from undersampled data with more details.
problem Recovering fine details from undersampled MRI data.
method Pyramid Convolutional RNN (PC-RNN) with three ConvRNN modules for multi-scale reconstruction.
result PC-RNN outperforms other methods in recovering more details from MRI images.
This paper combines deep learning with medical imaging models to improve reconstruction speed and reduce radiation.
problem Medical imaging issues like slow MRI, radiation injury in CT and PET.
method Combining deep learning with model-based reconstruction.
result Improves reconstruction speed and reduces radiation in medical imaging.
Algorithms for Magnetic Resonance (MR) image reconstruction from undersampled measurements exploit prior information to compensate for missing k-space data. Deep learning (DL) provides a powerful framework for extracting such information from existing image datasets, through learning, and then using it for reconstructi…
SUPER learning combines supervised and unsupervised methods for LDCT image reconstruction.
problem Low-dose CT image reconstruction challenges.
method Combines supervised and unsupervised learning methods.
result SUPER learning dramatically outperforms constituent methods.
The development of computed tomography (CT) image reconstruction methods that significantly reduce patient radiation exposure while maintaining high image quality is an important area of research in low-dose CT (LDCT) imaging. We propose a new penalized weighted least squares (PWLS) reconstruction method that exploits …
Enhances MRI image quality with Conditional WGAN and adaptive balancing.
problem Struggles to reconstruct sharp images with fine detail.
method Conditional Wasserstein Generative Adversarial Network (WGAN) with Adaptive Gradient Balancing.
result Produces sharper images than other techniques.
Sparsity and low-rank models have been popular for reconstructing images and videos from limited or corrupted measurements. Dictionary or transform learning methods are useful in applications such as denoising, inpainting, and medical image reconstruction. This paper proposes a framework for online (or time-sequential)…
A new deep learning model speeds up MRI by reconstructing from undersampled data.
problem Slow MRI due to undersampling in k-space.
method Unrolling primal-dual hybrid gradient algorithm into a deep network, gradually relaxing constraints.
result Superior MR reconstructions from highly undersampled data.
This study uses CNN-IOs to estimate MRI image reconstruction performance bounds.
problem Estimating task-based performance limits for MRI image reconstruction methods.
method Utilized stylized multi-coil SENSE MRI systems and deep-generated stochastic models to estimate IO performance.
result Estimation of IO performance provides guidance for designing under-sampled MRI systems.
Study examines stability of image-reconstruction algorithms using variational regularization.
problem Stability and robustness of image-reconstruction algorithms in medical imaging.
method Review and novel stability results for ℓp-regularized linear inverse problems, focusing on p∈(1,∞). result Guarantees Lipschitz continuity for small p and Hölder continuity for larger p in Lp(Ω) function spaces. Efficient deep learning for CT images reduces memory and training time.
problem Training deep neural networks for CT images is computationally expensive.
method Unrolled proximal gradient descent, replaced penalty terms with CNNs, used greedy learning with deep UNet and surrogate.
result Achieved comparable image quality to state-of-the-art methods on CT image reconstruction challenges.
Adversarial perturbations are more effective in Y-channel of YCbCr color space.
problem Vulnerability of deep models to adversarial perturbations in images.
method Proposed ResUpNet defense that removes perturbations only from the Y-channel of YCbCr color space.
result ResUpNet achieves the best balance between defense and maintaining original image accuracy.
A fast method approximates likelihood scores for noisy linear inverse problems.
problem Solving noisy linear inverse problems efficiently.
method Proposes a simple closed-form approximation to the likelihood score for diffusion and flow-based models.
result Significantly faster than baseline methods while maintaining competitive or better reconstruction performances.
SaR-SVM-STV improves hyperspectral image classification with shape-adaptive reconstruction and denoising.
problem Classifying hyperspectral images with limited labeled data.
method Shape-adaptive Reconstruction (SaR) for pixel preprocessing, SVM for probability estimation, and Smoothed Total Variation (STV) for denoising.
result SaR-SVM-STV outperforms SVM-STV with fewer labeled data.
We discuss recently emerging applications of the state-of-art deep learning methods on optical microscopy and microscopic image reconstruction, which enable new transformations among different modes and modalities of microscopic imaging, driven entirely by image data. We believe that deep learning will fundamentally ch…
We consider the problem of reconstructing signals and images from periodic nonlinearities. For such problems, we design a measurement scheme that supports efficient reconstruction; moreover, our method can be adapted to extend to compressive sensing-based signal and image acquisition systems. Our techniques can be pote…