Unsupervised Attention-guided Image to Image Translation

Abstract

Current unsupervised image-to-image translation techniques struggle to focus their attention on individual objects without altering the background or the way multiple objects interact within a scene. Motivated by the important role of attention in human perception, we tackle this limitation by introducing unsupervised attention mechanisms that are jointly adversarialy trained with the generators and discriminators. We demonstrate qualitatively and quantitatively that our approach is able to attend to relevant regions in the image without requiring supervision, and that by doing so it achieves more realistic mappings compared to recent approaches.

Cited by in corpus (63)

• A General Survey on Attention Mechanisms in Deep Learning
• U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
• Improving the Harmony of the Composite Image by Spatial-Separated Attention Module
• InstaGAN: Instance-aware Image-to-Image Translation
• PSIGAN: Joint probabilistic segmentation and image distribution matching for unpaired cross-modality adaptation based MRI segmentation
• SMAPGAN: Generative Adversarial Network Based Semi-Supervised Styled Map Tiles Generating Method
• Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications
• Text as Neural Operator: Image Manipulation by Text Instruction
• Multi-Channel Attention Selection GAN with Cascaded Semantic Guidance for Cross-View Image Translation
• Weather GAN: Multi-Domain Weather Translation Using Generative Adversarial Networks
• Towards Visually Explaining Variational Autoencoders
• Image-to-Image Translation: Methods and Applications
• End-to-End Learning of Geometric Deformations of Feature Maps for Virtual Try-On
• Non-uniform Motion Deblurring with Blurry Component Divided Guidance
• Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping
• Attention-Guided Generative Adversarial Networks for Unsupervised Image-to-Image Translation
• Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation
• Anisotropic Stroke Control for Multiple Artists Style Transfer
• SAG-GAN: Semi-Supervised Attention-Guided GANs for Data Augmentation on Medical Images
• FREA-Unet: Frequency-aware U-net for Modality Transfer
• Cross-Domain Cascaded Deep Feature Translation
• MixerGAN: An MLP-Based Architecture for Unpaired Image-to-Image Translation
• The Spatially-Correlative Loss for Various Image Translation Tasks
• Image-to-image Translation via Hierarchical Style Disentanglement
• Cycle In Cycle Generative Adversarial Networks for Keypoint-Guided Image Generation
• Multi-Mapping Image-to-Image Translation with Central Biasing Normalization
• Describe What to Change: A Text-guided Unsupervised Image-to-Image Translation Approach
• Semantics-Aware Image to Image Translation and Domain Transfer
• Mocycle-GAN: Unpaired Video-to-Video Translation
• A Shared Representation for Photorealistic Driving Simulators
• Exploring Explicit Domain Supervision for Latent Space Disentanglement in Unpaired Image-to-Image Translation
• Dual-Attention GAN for Large-Pose Face Frontalization
• Mask Based Unsupervised Content Transfer
• EDIT: Exemplar-Domain Aware Image-to-Image Translation
• Unpaired Image-to-Image Translation using Adversarial Consistency Loss
• Dual Attention GANs for Semantic Image Synthesis
• Improving Noise Robustness In Speaker Identification Using A Two-Stage Attention Model
• Attention-Guided Generative Adversarial Network to Address Atypical Anatomy in Modality Transfer
• Global and Local Alignment Networks for Unpaired Image-to-Image Translation
• How Old Are You? Face Age Translation with Identity Preservation Using GANs
• Towards Lifelong Self-Supervision For Unpaired Image-to-Image Translation
• Unselfie: Translating Selfies to Neutral-pose Portraits in the Wild
• One-to-one Mapping for Unpaired Image-to-image Translation
• Semantically Robust Unpaired Image Translation for Data with Unmatched Semantics Statistics
• Person-in-Context Synthesiswith Compositional Structural Space
• Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection
• A Strong Feature Representation for Siamese Network Tracker
• TransferI2I: Transfer Learning for Image-to-Image Translation from Small Datasets
• Attention-based Adversarial Appearance Learning of Augmented Pedestrians
• A Novel Framework for Image-to-image Translation and Image Compression
• Stay Positive: Non-Negative Image Synthesis for Augmented Reality
• Synthesizing Photorealistic Images with Deep Generative Learning
• Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translation
• A Survey on Adversarial Image Synthesis
• Unaligned Image-to-Image Translation by Learning to Reweight
• Separating Content and Style for Unsupervised Image-to-Image Translation
• Cosmetic-Aware Makeup Cleanser
• What and Where to Translate: Local Mask-based Image-to-Image Translation
• Langevin Cooling for Domain Translation
• Generating Object Stamps
• Joint Wasserstein Distribution Matching
• Expression Conditional GAN for Facial Expression-to-Expression Translation
• From Caesar Cipher to Unsupervised Learning: A New Method for Classifier Parameter Estimation