Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:IRKS6MQRrecord.jsonopen to challenge →
read the original abstract
Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs. However, for many tasks, paired training data will not be available. We present an approach for learning to translate an image from a source domain $X$ to a target domain $Y$ in the absence of paired examples. Our goal is to learn a mapping $G: X \rightarrow Y$ such that the distribution of images from $G(X)$ is indistinguishable from the distribution $Y$ using an adversarial loss. Because this mapping is highly under-constrained, we couple it with an inverse mapping $F: Y \rightarrow X$ and introduce a cycle consistency loss to push $F(G(X)) \approx X$ (and vice versa). Qualitative results are presented on several tasks where paired training data does not exist, including collection style transfer, object transfiguration, season transfer, photo enhancement, etc. Quantitative comparisons against several prior methods demonstrate the superiority of our approach.
This paper has not been read by Pith yet.
Forward citations
Cited by 18 Pith papers
-
AGAN: Towards Automated Design of Generative Adversarial Networks
AGAN is the first neural architecture search method for GANs that discovers architectures outperforming state-of-the-art on CIFAR-10 unsupervised image generation and competitive on supervised tasks.
-
SyMTRS: Benchmark Multi-Task Synthetic Dataset for Depth, Domain Adaptation and Super-Resolution in Aerial Imagery
A new large-scale synthetic multi-task benchmark dataset supplying pixel-perfect depth, domain-shifted night imagery, and multi-scale low-resolution pairs for aerial remote sensing.
-
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Progressive growing stabilizes GAN training to produce high-resolution images of unprecedented quality and achieves a record unsupervised inception score of 8.80 on CIFAR10.
-
Multispectral Blind Image Super-Resolution for Standing Dead Tree Segmentation
A blind super-resolution method with attention-guided domain adaptation improves multispectral image quality for dead tree segmentation using unpaired data, achieving Dice scores of 54-64%.
-
Leveraging Text Repetitions and Denoising Autoencoders in OCR Post-correction
Error distributions estimated from text repetitions enable training of denoising autoencoders that improve OCR post-correction on historical Finnish newspapers without manual training data.
-
Adversarial Computation of Optimal Transport Maps
A GAN with Wasserstein discriminator objective makes the generator follow the W2 geodesic to learn an optimal transport map.
-
Demystifying MMD GANs
MMD GANs have unbiased critic gradients but biased generator gradients from sample-based learning, and the Kernel Inception Distance provides a practical new measure for GAN convergence and dynamic learning rate adaptation.
-
Unsupervised Susceptibility Distortion Correction of EPI without Calibration Scans via Image Translation-Based Registration
SACRED performs unsupervised susceptibility distortion correction of EPI fMRI via image translation-based registration between T1w and unidirectional BOLD images, with test-time adaptation for robustness.
-
Hist2Style: Histogram-Guided Stylization with Bilateral Grids
Hist2Style introduces a lightweight bilateral-grid network conditioned on histogram embeddings for distilling large-model stylization into real-time, structure-preserving, user-controllable photorealistic edits.
-
On the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systems
Experiments on real industrial time series show that partial model sharing improves diffusion model performance in bandwidth-limited non-IID settings, while full sharing stabilizes GAN training but offers less robustn...
-
Style-Based Neural Architectures for Real-Time Weather Classification
Three style-based neural architectures are proposed for real-time weather classification from images, with two truncated ResNet variants claimed to outperform prior methods and generalize across public datasets.
-
Stylistic-STORM (ST-STORM) : Perceiving the Semantic Nature of Appearance
ST-STORM introduces a dual-branch SSL framework that disentangles semantic content from stylistic appearance using gated latent streams, JEPA for content invariance, and adversarial constraints for style capture.
-
Heuristic Style Transfer for Real-Time, Efficient Weather Attribute Detection
Lightweight multi-task models using Gram matrices and PatchGAN-style architectures detect 53 weather classes from RGB images with F1 scores above 96% internally and 78% zero-shot externally, supported by a new 503k-im...
-
Multivariate-Information Adversarial Ensemble for Scalable Joint Distribution Matching
MMI-ALI extends pairwise ALI models into an m-domain ensemble by maximizing MMI on joint variables to achieve scalable joint distribution matching with linear scaling in m.
-
Disentangled Makeup Transfer with Generative Adversarial Network
DMT uses identity and makeup encoders in a GAN to enable controllable makeup transfer from references and sampling of new styles from a prior distribution.
-
Lightweight Cross-Spectral Face Recognition via Contrastive Alignment and Distillation
A lightweight hybrid CNN-Transformer framework for heterogeneous face recognition achieves competitive performance on cross-spectral benchmarks and standard RGB tasks using contrastive alignment and distillation.
-
SAGE-GAN: Towards Realistic and Robust Segmentation of Spatially Ordered Nanoparticles via Attention-Guided GANs
SAGE-GAN integrates a self-attention U-Net into a CycleGAN framework to generate realistic synthetic electron microscopy image-mask pairs that augment training data for nanoparticle segmentation without human labeling.
-
Hierarchical Sequence to Sequence Voice Conversion with Limited Data
Hierarchical seq2seq model for parallel voice conversion pretrained as autoencoder on single-speaker data then adapted to limited multispeaker data, using mel spectrograms converted via wavenet vocoder.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.