Methods › General › Loss Functions › Cycle Consistency Loss
Cycle Consistency Loss
Introduced by Jun-Yan Zhu et al. in Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Cycle Consistency Loss is a type of loss used for generative adversarial networks that performs unpaired image-to-image translation. It was introduced with the CycleGAN architecture. For two domains X and Y, we want to learn a mapping G : X →Y and F: Y →X. We want to enforce the intuition that these mappings should be reverses of each other and that both mappings should be bijections. Cycle Consistency Loss encourages F(G(x)) ≈x and G(F(y)) ≈y. It reduces the space of possible mapping functions by enforcing forward and backwards consistency:
ℒ_(cyc)(G, F) = 𝔼_(x ∼p_(data)(x))[||F(G(x)) - x||₁] + 𝔼_(y ∼p_(data)(y))[||G(F(y)) - y||₁]
In Syntology for a different name, 'cycle-consistency', about another paper, arXiv:2108.03427: repository known, no verified implementation (declared_code; repositories: 2 listed of 2 in the graph, strongest link: official). The lookup matched a different name in Syntology's graph, 'cycle-consistency' (resolution: token_match); this answer is about that name. About another paper: this answer is about arXiv:2108.03427, while the archive credits arXiv:1703.10593.
Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.
- consistency models · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='consistency models') - consistency policy · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='consistency policy') - self-consistency loss · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='self-consistency loss') - Self-Consistency Training · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Self-Consistency Training') - self-consistency approach · shares_name_tokens (shares name tokens with this name; a different method) · level V2 · retrieve with
get_reference_implementation(name='self-consistency approach') - Direct Consistency Optimization · shares_name_tokens (shares name tokens with this name; a different method) · level V2 · retrieve with
get_reference_implementation(name='Direct Consistency Optimization')
Origin paper of this answer, arXiv:2108.03427 (its page on this site · its page on Syntology): Syntology's run record for its harvested code, 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified (of 2 harvested samples) · 2 pointer-only (licence). The instrument figure counts failures of Syntology's instrument, not of the code. Ran means executed on a synthesized input, not that the code is correct.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
What each level means, quoted from Syntology's server card: V0 “imports”; V1 “executes against a synthesized fixture and returns a value of the recorded type”; V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” The card adds: “No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results.” A level quoted in two parts is one definition the card writes as two clauses joined by a double hyphen, each clause quoted on its own (how these levels are quoted).
Papers archive 2025-07-28
archive papers tagged: 448 · with a code link: 148 · where Syntology ran a sample: 12 (9 with a run with no instrument failure, 3 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (12 of 448 tagged: 9 with a run with no instrument failure, 3 where every run was a failure of Syntology's instrument)
Page 1 of 5: papers 1 to 100 of 448, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
Multipath cycleGAN for harmonization of paired and unpaired low-dose lung computed tomography reconstruction kernels 28 May 2025 · 0 repositories · arXiv:2505.22568
-
Unpaired Image-to-Image Translation for Segmentation and Signal Unmixing 27 May 2025 · 0 repositories · arXiv:2505.20746
-
3D Reconstruction from Sketches 20 May 2025 · 0 repositories · arXiv:2505.14621
-
Towards Generating Realistic Underwater Images 20 May 2025 · 0 repositories · arXiv:2505.14296
-
A Deep Learning-Driven Inhalation Injury Grading Assistant Using Bronchoscopy Images 13 May 2025 · 0 repositories · arXiv:2505.08517
-
MRI motion correction via efficient residual-guided denoising diffusion probabilistic models 6 May 2025 · 0 repositories · arXiv:2505.03498
-
Revolutionizing Brain Tumor Imaging: Generating Synthetic 3D FA Maps from T1-Weighted MRI using CycleGAN Models 6 May 2025 · 0 repositories · arXiv:2505.03662
-
Lesion-Aware Generative Artificial Intelligence for Virtual Contrast-Enhanced Mammography in Breast Cancer 5 May 2025 · 0 repositories · arXiv:2505.03018
-
ClearVision: Leveraging CycleGAN and SigLIP-2 for Robust All-Weather Classification in Traffic Camera Imagery 28 Apr 2025 · 0 repositories · arXiv:2504.19684
-
Test-Time Visual In-Context Tuning 27 Mar 2025 · 1 repository · arXiv:2503.21777
-
Tune It Up: Music Genre Transfer and Prediction 27 Mar 2025 · 1 repository · arXiv:2503.22008
-
Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training 19 Mar 2025 · 1 repository · arXiv:2503.15017
-
Whole-Body Image-to-Image Translation for a Virtual Scanner in a Healthcare Digital Twin 18 Mar 2025 · 0 repositories · arXiv:2503.15555
-
CyclePose -- Leveraging Cycle-Consistency for Annotation-Free Nuclei Segmentation in Fluorescence Microscopy 14 Mar 2025 · 0 repositories · arXiv:2503.11266
-
Feature Fusion Attention Network with CycleGAN for Image Dehazing, De-Snowing and De-Raining 8 Mar 2025 · 0 repositories · arXiv:2503.06107
-
Generalizable Image Repair for Robust Visual Autonomous Racing 7 Mar 2025 · 1 repository · arXiv:2503.05911
-
MRI super-resolution reconstruction using efficient diffusion probabilistic model with residual shifting 3 Mar 2025 · 1 repository · arXiv:2503.01576
-
Cross Modality Medical Image Synthesis for Improving Liver Segmentation 2 Mar 2025 · 0 repositories · arXiv:2503.00945
-
Development of an Unpaired Deep Neural Network for Synthesizing X-ray Fluoroscopic Images from Digitally Reconstructed Tomography in Image Guided Radiotherapy 1 Mar 2025 · 0 repositories · arXiv:2503.00665
-
Single-image Reflectance and Transmittance Estimation from Any Flatbed Scanner 20 Feb 2025 · 0 repositories · arXiv:2502.14462
-
A Physics-Informed Deep Learning Model for MRI Brain Motion Correction 13 Feb 2025 · 1 repository · arXiv:2502.09296
-
Bidirectional Brain Image Translation using Transfer Learning from Generic Pre-trained Models 21 Jan 2025 · 0 repositories · arXiv:2501.12488
-
SFC-GAN: A Generative Adversarial Network for Brain Functional and Structural Connectome Translation 13 Jan 2025 · 0 repositories · arXiv:2501.07055
-
A Value Mapping Virtual Staining Framework for Large-scale Histological Imaging 7 Jan 2025 · 0 repositories · arXiv:2501.03592
-
Plasma-CycleGAN: Plasma Biomarker-Guided MRI to PET Cross-modality Translation Using Conditional CycleGAN 4 Jan 2025 · 0 repositories · arXiv:2501.02146
-
Emulating Clinical Quality Muscle B-mode Ultrasound Images from Plane Wave Images Using a Two-Stage Machine Learning Model 7 Dec 2024 · 1 repository · arXiv:2412.05758
-
Improving speaker verification robustness with synthetic emotional utterances 30 Nov 2024 · 0 repositories · arXiv:2412.00319
-
Improving Deformable Image Registration Accuracy through a Hybrid Similarity Metric and CycleGAN Based Auto-Segmentation 25 Nov 2024 · 0 repositories · arXiv:2411.16992
-
Unified Microphone Conversion: Many-to-Many Device Mapping via Feature-wise Linear Modulation 23 Oct 2024 · 0 repositories · arXiv:2410.18322
-
Transesophageal Echocardiography Generation using Anatomical Models 9 Oct 2024 · 0 repositories · arXiv:2410.06781
-
Training a Computer Vision Model for Commercial Bakeries with Primarily Synthetic Images 30 Sep 2024 · 0 repositories · arXiv:2409.20122
-
Enhancing Synthetic Training Data for Speech Commands: From ASR-Based Filtering to Domain Adaptation in SSL Latent Space 19 Sep 2024 · 0 repositories · arXiv:2409.12745
-
I2I-Galip: Unsupervised Medical Image Translation Using Generative Adversarial CLIP 19 Sep 2024 · 1 repository · arXiv:2409.12399
-
Improving Cone-Beam CT Image Quality with Knowledge Distillation-Enhanced Diffusion Model in Imbalanced Data Settings 19 Sep 2024 · 0 repositories · arXiv:2409.12539
-
Schrödinger Bridge Flow for Unpaired Data Translation 14 Sep 2024 · 0 repositories · arXiv:2409.09347
-
CONDA: Condensed Deep Association Learning for Co-Salient Object Detection 2 Sep 2024 · 0 repositories · arXiv:2409.01021
-
Seed-to-Seed: Image Translation in Diffusion Seed Space 1 Sep 2024 · 0 repositories · arXiv:2409.00654
-
CycleGAN with Better Cycles 27 Aug 2024 · 0 repositories · arXiv:2408.15374
-
Bridging the Gap between Real-world and Synthetic Images for Testing Autonomous Driving Systems 25 Aug 2024 · 0 repositories · arXiv:2408.13950
-
JieHua Paintings Style Feature Extracting Model using Stable Diffusion with ControlNet 21 Aug 2024 · 0 repositories · arXiv:2408.11744
-
FQGA-single: Towards Fewer Training Epochs and Fewer Model Parameters for Image-to-Image Translation Tasks 17 Aug 2024 · 0 repositories · arXiv:2408.09218
-
The Quest for Early Detection of Retinal Disease: 3D CycleGAN-based Translation of Optical Coherence Tomography into Confocal Microscopy 7 Aug 2024 · 0 repositories · arXiv:2408.04091
-
BodySLAM: A Generalized Monocular Visual SLAM Framework for Surgical Applications 6 Aug 2024 · 1 repository · arXiv:2408.03078
-
Improving noisy student training for low-resource languages in End-to-End ASR using CycleGAN and inter-domain losses 26 Jul 2024 · 0 repositories · arXiv:2407.21061
-
Utilizing Generative Adversarial Networks for Image Data Augmentation and Classification of Semiconductor Wafer Dicing Induced Defects 24 Jul 2024 · 0 repositories · arXiv:2407.20268
-
CycleMix: Mixing Source Domains for Domain Generalization in Style-Dependent Data 18 Jul 2024 · 1 repository · arXiv:2407.13421
-
CycleHOI: Improving Human-Object Interaction Detection with Cycle Consistency of Detection and Generation 16 Jul 2024 · 0 repositories · arXiv:2407.11433
-
Theoretical Insights into CycleGAN: Analyzing Approximation and Estimation Errors in Unpaired Data Generation 16 Jul 2024 · 0 repositories · arXiv:2407.11678
-
Leveraging Multimodal CycleGAN for the Generation of Anatomically Accurate Synthetic CT Scans from MRIs 15 Jul 2024 · 0 repositories · arXiv:2407.10888
-
HSFusion: A high-level vision task-driven infrared and visible image fusion network via semantic and geometric domain transformation 14 Jul 2024 · 0 repositories · arXiv:2407.10047
-
Deep-Motion-Net: GNN-based volumetric organ shape reconstruction from single-view 2D projections 9 Jul 2024 · 0 repositories · arXiv:2407.06692
-
A Domain Adaptation Model for Carotid Ultrasound: Image Harmonization, Noise Reduction, and Impact on Cardiovascular Risk Markers 6 Jul 2024 · 0 repositories · arXiv:2407.05163
-
Cyclic 2.5D Perceptual Loss for Cross-Modal 3D Medical Image Synthesis: T1w MRI to Tau PET 18 Jun 2024 · 1 repository · arXiv:2406.12632
-
CORE-ReID: Comprehensive Optimization and Refinement through Ensemble Fusion in Domain Adaptation for Person Re-Identification 3 Jun 2024 · 1 repository
-
S-CycleGAN: Semantic Segmentation Enhanced CT-Ultrasound Image-to-Image Translation for Robotic Ultrasonography 3 Jun 2024 · 1 repository · arXiv:2406.01191
-
Cyclic image generation using chaotic dynamics 31 May 2024 · 1 repository · arXiv:2405.20717
-
Cycle-YOLO: A Efficient and Robust Framework for Pavement Damage Detection 28 May 2024 · 0 repositories · arXiv:2405.17905
-
Shape-aware synthesis of pathological lung CT scans using CycleGAN for enhanced semi-supervised lung segmentation 14 May 2024 · 1 repository · arXiv:2405.08556
-
PotatoGANs: Utilizing Generative Adversarial Networks, Instance Segmentation, and Explainable AI for Enhanced Potato Disease Identification and Classification 12 May 2024 · 1 repository · arXiv:2405.07332
-
Domain-Transferred Synthetic Data Generation for Improving Monocular Depth Estimation 2 May 2024 · 0 repositories · arXiv:2405.01113
-
MeshBrush: Painting the Anatomical Mesh with Neural Stylization for Endoscopy 3 Apr 2024 · 1 repository · arXiv:2404.02999
-
StainDiffuser: MultiTask Dual Diffusion Model for Virtual Staining 17 Mar 2024 · 0 repositories · arXiv:2403.11340
-
Could We Generate Cytology Images from Histopathology Images? An Empirical Study 16 Mar 2024 · 0 repositories · arXiv:2403.10885
-
Scene Depth Estimation from Traditional Oriental Landscape Paintings 6 Mar 2024 · 0 repositories · arXiv:2403.03408
-
Cross-Domain Image Conversion by CycleDM 5 Mar 2024 · 0 repositories · arXiv:2403.02919
-
(Un)paired signal-to-signal translation with 1D conditional GANs 5 Mar 2024 · 0 repositories · arXiv:2403.04800
-
Learning to Find Missing Video Frames with Synthetic Data Augmentation: A General Framework and Application in Generating Thermal Images Using RGB Cameras 29 Feb 2024 · 0 repositories · arXiv:2403.00196
-
Reviewing FID and SID Metrics on Generative Adversarial Networks 6 Feb 2024 · 0 repositories · arXiv:2402.03654
-
IGUANe: a 3D generalizable CycleGAN for multicenter harmonization of brain MR images 5 Feb 2024 · 1 repository · arXiv:2402.03227
-
Unsupervised Generation of Pseudo Normal PET from MRI with Diffusion Model for Epileptic Focus Localization 2 Feb 2024 · 0 repositories · arXiv:2402.01191
-
High-Quality Medical Image Generation from Free-hand Sketch 1 Feb 2024 · 0 repositories · arXiv:2402.00353
-
CIMGEN: Controlled Image Manipulation by Finetuning Pretrained Generative Models on Limited Data 23 Jan 2024 · 0 repositories · arXiv:2401.13006
-
Grayscale Image Colorization with GAN and CycleGAN in Different Image Domain 21 Jan 2024 · 0 repositories · arXiv:2401.11425
-
Thermal Image Calibration and Correction using Unpaired Cycle-Consistent Adversarial Networks 21 Jan 2024 · 0 repositories · arXiv:2401.11582
-
Towards Identifiable Unsupervised Domain Translation: A Diversified Distribution Matching Approach 18 Jan 2024 · 0 repositories · arXiv:2401.09671
-
Microphone Conversion: Mitigating Device Variability in Sound Event Classification 12 Jan 2024 · 1 repository · arXiv:2401.06913
-
Evaluating Data Augmentation Techniques for Coffee Leaf Disease Classification 11 Jan 2024 · 0 repositories · arXiv:2401.05768
-
Evaluation of Mean Shift, ComBat, and CycleGAN for Harmonizing Brain Connectivity Matrices Across Sites 9 Jan 2024 · 0 repositories · arXiv:2401.06798
-
CycleGAN Models for MRI Image Translation 28 Dec 2023 · 0 repositories · arXiv:2401.00023
-
Compressing Image-to-Image Translation GANs Using Local Density Structures on Their Learned Manifold 22 Dec 2023 · 0 repositories · arXiv:2312.14776
-
Ultrasound Image Enhancement using CycleGAN and Perceptual Loss 18 Dec 2023 · 1 repository · arXiv:2312.11748
-
PhenDiff: Revealing Subtle Phenotypes with Diffusion Models in Real Images 13 Dec 2023 · 1 repository · arXiv:2312.08290Syntology official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified (of 8 harvested samples) · 8 pointer-only (licence)
-
Editable Stain Transformation Of Histological Images Using Unpaired GANs 6 Dec 2023 · 1 repository · arXiv:2312.03647
-
Pipeline Enabling Zero-shot Classification for Bangla Handwritten Grapheme 1 Dec 2023 · 0 repositories
-
Towards ultra-low-cost smartphone microscopy 28 Nov 2023 · 0 repositories · arXiv:2312.11479
-
OCT2Confocal: 3D CycleGAN based Translation of Retinal OCT Images to Confocal Microscopy 17 Nov 2023 · 2 repositories · arXiv:2311.10902
-
A Novel 1D Generative Adversarial Network-based Framework for Atrial Fibrillation Detection using Restored Wrist Photoplethysmography Signals 13 Nov 2023 · 0 repositories · arXiv:2312.09459
-
CycleGANAS: Differentiable Neural Architecture Search for CycleGAN 13 Nov 2023 · 1 repository · arXiv:2311.07162
-
Synthesizing Bidirectional Temporal States of Knee Osteoarthritis Radiographs with Cycle-Consistent Generative Adversarial Neural Networks 10 Nov 2023 · 0 repositories · arXiv:2311.05798
-
Let's Get the FACS Straight -- Reconstructing Obstructed Facial Features 9 Nov 2023 · 0 repositories · arXiv:2311.05221
-
A Two-Stage Generative Model with CycleGAN and Joint Diffusion for MRI-based Brain Tumor Detection 6 Nov 2023 · 1 repository · arXiv:2311.03074
-
Deep Image Semantic Communication Model for Artificial Intelligent Internet of Things 6 Nov 2023 · 2 repositories · arXiv:2311.02926
-
Novel View Synthesis from a Single RGBD Image for Indoor Scenes 2 Nov 2023 · 0 repositories · arXiv:2311.01065
-
Feature-oriented Deep Learning Framework for Pulmonary Cone-beam CT (CBCT) Enhancement with Multi-task Customized Perceptual Loss 1 Nov 2023 · 1 repository · arXiv:2311.00412
-
DT/MARS-CycleGAN: Improved Object Detection for MARS Phenotyping Robot 19 Oct 2023 · 0 repositories · arXiv:2310.12787
-
LROC-PANGU-GAN: Closing the Simulation Gap in Learning Crater Segmentation with Planetary Simulators 4 Oct 2023 · 0 repositories · arXiv:2310.02781
-
Prompt-based test-time real image dehazing: a novel pipeline 29 Sep 2023 · 1 repository · arXiv:2309.17389
-
Stochastic Digital Twin for Copy Detection Patterns 28 Sep 2023 · 0 repositories · arXiv:2309.16866
-
A Novel Deep Learning Technique for Morphology Preserved Fetal ECG Extraction from Mother ECG using 1D-CycleGAN 25 Sep 2023 · 0 repositories · arXiv:2310.03759
-
OSNet & MNetO: Two Types of General Reconstruction Architectures for Linear Computed Tomography in Multi-Scenarios 21 Sep 2023 · 0 repositories · arXiv:2309.11858
Tasks archive 2025-07-28
20 shown of 306 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Translation | 112 |
| Image-to-Image Translation | 92 |
| Image Generation | 50 |
| Generative Adversarial Network | 49 |
| Domain Adaptation | 42 |
| Data Augmentation | 33 |
| Semantic Segmentation | 30 |
| Style Transfer | 30 |
| Segmentation | 28 |
| SSIM | 26 |
| Computed Tomography (CT) | 16 |
| Diagnostic | 15 |
| Object Detection | 15 |
| Unsupervised Domain Adaptation | 15 |
| object-detection | 15 |
| Denoising | 14 |
| Super-Resolution | 14 |
| Anatomy | 13 |
| Image Segmentation | 13 |
| Diversity | 11 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections