IntroVAE synthesizes high-quality photos by self-evaluating and improving its outputs.
problem Generating high-quality photographic images with stable and realistic results.
method Introspective Variational Autoencoder (IntroVAE) that trains inference and generator models jointly, encouraging the inference model to distinguish between generated and real samples.
result Produces high-resolution photo-realistic images comparable to or better than state-of-the-art GANs.
Study uses CNNs to estimate BMI from photos.
problem Estimating BMI from photos with limited data.
method Used deep learning (CNNs) to process silhouette images.
result High correlation between estimated and actual BMI.
We present a method for synthesizing a frontal, neutral-expression image of a person's face given an input face photograph. This is achieved by learning to generate facial landmarks and textures from features extracted from a facial-recognition network. Unlike previous approaches, our encoding feature vector is largely…
The modulation transfer function (MTF) is widely used to characterise the performance of optical systems. Measuring it is costly and it is thus rarely available for a given lens specimen. Instead, MTFs based on simulations or, at best, MTFs measured on other specimens of the same lens are used. Fortunately, images reco…
HiLLoC compresses large images losslessly using VAEs.
problem Lossless compression of large color photographs.
method Fully convolutional VAE models trained on ImageNet are applied to lossless compression.
result Achieves state-of-the-art compression for full-size ImageNet images.
We propose a new algorithm for training generative adversarial networks that jointly learns latent codes for both identities (e.g. individual humans) and observations (e.g. specific photographs). By fixing the identity portion of the latent codes, we can generate diverse images of the same subject, and by fixing the ob…
Framework synthesizes geological images minimizing patch distribution discrepancy.
problem Synthesizing realistic geological images from a single exemplar.
method Uses kernel discrepancies and generative neural networks for efficient synthesis.
result Synthesized images match visual patterns and spatial statistics of the exemplar.
New method evaluates text-to-image synthesis for realism, variety, and semantic accuracy.
problem Lack of metrics revealing semantic accuracy in text-to-image synthesis.
method Uses Inception network representations and t-SNE visualization for semantic evaluation.
result Classification accuracy of generated images to real images' visual concepts correlates with semantic accuracy.
We introduce a new dataset of 293,008 high definition (1360 x 1360 pixels) fashion images paired with item descriptions provided by professional stylists. Each item is photographed from a variety of angles. We provide baseline results on 1) high-resolution image generation, and 2) image generation conditioned on the gi…
We apply basic statistical reasoning to signal reconstruction by machine learning -- learning to map corrupted observations to clean signals -- with a simple and powerful conclusion: it is possible to learn to restore images by only looking at corrupted examples, at performance at and sometimes exceeding training using…
Deep learning predicts diabetic macular edema from fundus photos.
problem Diabetic macular edema diagnosis from fundus photos is inaccurate.
method Trained deep learning model on color fundus photographs.
result Deep learning model has higher sensitivity and PPV than human specialists.
Diffusion models outperform GANs in image synthesis quality.
problem Improving image synthesis quality using generative models.
method Developed and optimized diffusion models, and introduced classifier guidance for conditional synthesis.
result Achieved superior image synthesis quality with diffusion models, matching or exceeding GANs in FID scores.
Generative model creates realistic scenes from pixel-wise labels.
problem Creating photo-realistic scenes from pixel-wise labels.
method Semantic bottleneck GAN model combining conditional and unconditional generation networks.
result Model outperforms state-of-the-art models in unsupervised image synthesis.
GANs improve image generation from text and other images.
problem Improving image generation from text and other images.
method Adversarial training to estimate real data distribution and generate synthetic data.
result GANs outperform conventional methods in image generation.
Method treats pseudo healthy synthesis as a factor decomposition problem.
problem Creating a healthy-looking image from a pathological one.
method Adversarial training with paired or unpaired settings, combining two factors (healthy and disease) to reconstruct the input.
result Method outperforms conditional GAN and CycleGAN in generating pseudo healthy images.
A single robust classifier handles complex image synthesis tasks.
problem Challenging image synthesis tasks.
method Adversarial robustness training of a single classifier.
result Direct manipulation of salient features in input images.
This study shows how EEG can be used to generate fMRI data.
problem Mapping fMRI from EEG signals.
method Deep learning approaches (Autoencoders, GANs, Pairwise Learning).
result Feasibility of EEG to fMRI brain image mappings.
Single-stage neural architecture improves text-to-image synthesis.
problem Combining text and image generation is challenging.
method Used deep residual networks and sentence interpolation.
result Achieved state-of-the-art performance with single-stage training.
Generative model creates meal images from ingredient descriptions.
problem Synthesize photo-realistic meal images from ingredient descriptions.
method Attention-based ingredients-image association model, cycle-consistent constraint.
result Model generates meal images corresponding to ingredient descriptions.
Generative adversarial networks (GANs) are a recent approach to train generative models of data, which have been shown to work particularly well on image data. In the current paper we introduce a new model for texture synthesis based on GAN learning. By extending the input noise distribution space from a single vector …
Synthesizing high resolution photorealistic images has been a long-standing challenge in machine learning. In this paper we introduce new methods for the improved training of generative adversarial networks (GANs) for image synthesis. We construct a variant of GANs employing label conditioning that results in 128x128 r…
MetalGAN synthesizes images across multiple domains without labels.
problem Synthesizing images across multiple domains using a single network.
method Combines cGAN for image generation and Meta-Learning for domain switch.
result MetalGAN successfully produces multi-domain images without hard-coded labels.
BigGAN achieves state-of-the-art image synthesis on ImageNet.
problem Generating high-fidelity images from complex datasets like ImageNet.
method Trained a large-scale GAN with orthogonal regularization and a truncation trick.
result Improved Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4.
In this paper, we propose Generative Adversarial Network (GAN) architectures that use Capsule Networks for image-synthesis. Based on the principal of positional-equivariance of features, Capsule Network's ability to encode spatial relationships between the features of the image helps it become a more powerful critic in…
DreamFusion uses text-to-image diffusion models to create 3D images efficiently.
problem Lack of large-scale 3D datasets and efficient architectures for 3D synthesis.
method Adapting a 2D diffusion model to 3D synthesis using a loss based on probability density distillation.
result A 3D model can be optimized from a 2D diffusion model, allowing for text-to-3D synthesis.
A geometric account explains why 'The Dress' is ambiguous, predicting observable signatures in image processing.
problem Understanding and predicting ambiguity in image processing, particularly in intrinsic image decomposition.
method Geometric analysis of intrinsic image decomposition, focusing on the discontinuous switch in prior-mode sections.
result Predicted signatures in albedo Jacobian and Fernet curvature can be observed in various models and datasets.
Improved image synthesis with user scribbles and text prompts.
problem Inadequate details in generated images due to domain shift.
method Optimization problem formulation and cross-attention for control.
result Significant improvement in user satisfaction (85.32% higher).
Worldsheet wraps a 3D mesh sheet onto a single image to synthesize novel views.
problem Synthesizing novel views from a single image with large viewpoint changes.
method Shrink-wrapping a planar mesh sheet onto the input image, consistent with learned depth.
result Worldsheet consistently outperforms prior methods on single-image view synthesis.
This paper learns hierarchical compositional models for image synthesis.
problem Creating interpretable and hierarchical image representations.
method Sparsifying a generator network to induce an AND-OR hierarchy.
result The method learns meaningful and interpretable hierarchical representations.
Generative models learn complex spatial patterns using program synthesis.
problem Capturing complex global structure in data, especially in images.
method Incorporates programs representing global structure into generative models and learns these models through program synthesis.
result Significantly better at generating and completing images with global structure compared to state-of-the-art methods.
Program synthesis struggles with complex spatial relationships in image classification.
problem Challenges in solving Synthetic Visual Reasoning Test problems.
method Quantitative reanalysis of human and machine performance, improved program synthesis classifier, categorization of SVRT problems.
result Program synthesis is constrained by spatial relationships in images, not just shape specification.
Enhances VAEs for sharper image synthesis.
problem Blurriness in generated images from VAEs.
method Integrates a downscaled version of the original image into the VAE framework and uses it as input to the decoder.
result Improves FID score in image synthesis while maintaining similar log-likelihood performance.
Paper proposes method to generate images from text using GANs trained on uncaptioned images.
problem Limited captioned image datasets for text-to-image synthesis.
method Conditional GANs trained on uncaptioned images with an Image Captioning Module.
result Promising preliminary results compared to unconditional GANs.
Improved texture synthesis using wavelet-based statistics with rectifier non-linearity.
problem Improving texture synthesis quality using wavelet representations.
method Proposes a family of statistics based on non-linear wavelet representations with a generalized rectifier non-linearity.
result Significantly improves visual quality of texture synthesis compared to classical wavelet-based models.
Glow uses invertible 1x1 convolutions to improve image generation and manipulation.
problem Efficient and realistic image generation and manipulation.
method Invertible 1x1 convolutions in generative flows.
result Significant improvement in log-likelihood and realistic image synthesis.
A method to improve image synthesis diversity using mutual information.
problem Mode collapse in conditional GANs for multimodal image synthesis.
method Explicitly estimate and maximize mutual information between latent code and output image.
result Prevents mode collapse and encourages synthesis of diverse images.
Macrocanonical models generate textures matching input features.
problem Generating textures that match specific features.
method Use Gibbs measures and minimize a convex function for sampling.
result Macrocanonical models can be applied to real-valued images under certain conditions.
CollaGAN learns MR contrast imputation, revealing essential contrasts.
problem Understanding which MR contrast is essential for imputation.
method Collaborative Generative Adversarial Networks (CollaGAN).
result Exogenous contrast from contrast agents is not replaceable, but other endogenous contrasts can be synthesized.
Self-supervised GANs improve image synthesis without labeled data.
problem Lack of labeled data for conditional GANs.
method Combines adversarial training and self-supervision to learn meaningful feature representations.
result Self-supervised GANs achieve comparable performance to state-of-the-art conditional GANs.
High-quality image synthesis with diffusion models, achieving state-of-the-art FID score.
problem Generating high-quality images from latent variables.
method Training diffusion probabilistic models with a weighted variational bound, inspired by denoising score matching and Langevin dynamics.
result State-of-the-art FID score of 3.17 on CIFAR10 dataset.
Dataset of 50k hotels aids in human trafficking investigations.
problem Identify hotels from low-quality, occluded images.
method Curated dataset of 1M annotated hotel room images, baseline model with data augmentation.
result Demonstrated the feasibility of hotel recognition from challenging images.
Paper presents a GAN model for realistic river image synthesis.
problem Generating high-quality river images for hydrological research.
method Used a Progressive Growing GAN (PGGAN) architecture to overcome training challenges.
result Demonstrated the effectiveness of GANs in generating high-resolution river images.
The field of image classification has shown an outstanding success thanks to the development of deep learning techniques. Despite the great performance obtained, most of the work has focused on natural images ignoring other domains like artistic depictions. In this paper, we use transfer learning techniques to propose …
StyleNeRF generates high-resolution images with 3D consistency and style control.
problem Generating high-resolution images with fine details and 3D consistency.
method Integrates NeRF into a style-based generator for efficient high-resolution image synthesis.
result Synthesizes high-resolution images at interactive rates with high 3D consistency and style control.
C4Synth generates images from multiple captions to improve image quality.
problem Generating images from a single caption is limited; multiple captions are needed.
method Two deep generative models that ensure 'Cross-Caption Cycle Consistency'.
result Quantitative and qualitative validation on Caltech-UCSD Birds and Oxford-102 Flowers datasets.
Master thesis uses CNNs to estimate bike attractivity from street view images.
problem Estimating bike routes attractiveness from image data.
method Abstraction of locations into nodes and edges, enhancement with street view photos and neighborhood info, transfer learning.
result Improved road scoring using enhanced dataset and various model architectures.
Improved diffusion models for image synthesis with better training dynamics.
problem Uneven and ineffective training in diffusion models.
method Redesigned network layers to preserve activation, weight, and update magnitudes.
result Significantly better networks at equal computational complexity, improving FID to 1.81.
In this paper, we propose a novel application of Generative Adversarial Networks (GAN) to the synthesis of cells imaged by fluorescence microscopy. Compared to natural images, cells tend to have a simpler and more geometric global structure that facilitates image generation. However, the correlation between the spatial…