2018 CVPR GAN 相关论文调研 （自己分了下类，附地址哦）

2018 CVPR GAN 相关论文调研

风格迁移

1. PairedCycleGAN: Asymmetric Style Transfer for Applying and Removing Makeup

（给人脸化妆的风格转移）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Chang_PairedCycleGAN_Asymmetric_Style_CVPR_2018_paper.pdf

 

2.CartoonGAN: Generative Adversarial Networks for Photo Cartoonization

（将图片转化为卡通风格的GAN）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_CartoonGAN_Generative_Adversarial_CVPR_2018_paper.pdf

 

3.StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation

（人脸多种风格转换）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Choi_StarGAN_Unified_Generative_CVPR_2018_paper.pdf

 

4.Multi-Content GAN for Few-Shot Font Style Transfer

（字体风格转换）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Azadi_Multi-Content_GAN_for_CVPR_2018_paper.pdf

 

5.DA-GAN: Instance-level Image Translation by Deep Attention Generative Adversarial Networks

（图到图转换）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Ma_DA-GAN_Instance-Level_Image_CVPR_2018_paper.pdf

 

6. Conditional Image-to-Image translation

（图到图的转换）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Lin_Conditional_Image-to-Image_Translation_CVPR_2018_paper.pdf

 

图片处理

1. DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

（去模糊）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Kupyn_DeblurGAN_Blind_Motion_CVPR_2018_paper.pdf

 

2.Attentive Generative Adversarial Network for Raindrop Removal from A Single Image

（去除图片中的雨滴）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Qian_Attentive_Generative_Adversarial_CVPR_2018_paper.pdf

 

3. Deep Photo Enhancer: Unpaired Learning for Image Enhancement from Photographs with GANs

（用于照片增强）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_Deep_Photo_Enhancer_CVPR_2018_paper.pdf

 

4. SeGAN: Segmenting and Generating the Invisible

（去遮挡）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Ehsani_SeGAN_Segmenting_and_CVPR_2018_paper.pdf

 

5.Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal

（去阴影）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_Stacked_Conditional_Generative_CVPR_2018_paper.pdf

 

6.Image Blind Denoising With Generative Adversarial Network Based Noise Modeling

（去噪声）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_Image_Blind_Denoising_CVPR_2018_paper.pdf

 

7. Single Image Dehazing via Conditional Generative Adversarial Network

（去噪声）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Li_Single_Image_Dehazing_CVPR_2018_paper.pdf

 

图片生成

1. ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing

（空间转换生成图片）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Lin_ST-GAN_Spatial_Transformer_CVPR_2018_paper.pdf

 

2. SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis

（由边框生成图片）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_SketchyGAN_Towards_Diverse_CVPR_2018_paper.pdf

 

3. TextureGAN: Controlling Deep Image Synthesis with Texture Patches

（由纹路生成图片）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Xian_TextureGAN_Controlling_Deep_CVPR_2018_paper.pdf

 

4. Eye In-Painting with Exemplar Generative Adversarial Networks

（给人物画眼睛）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Dolhansky_Eye_In-Painting_With_CVPR_2018_paper.pdf

 

5.Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network

（文本生成图片）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Zhang_Photographic_Text-to-Image_Synthesis_CVPR_2018_paper.pdf

 

6. Logo Synthesis and Manipulation with Clustered Generative Adversarial Networks

（生成logo）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Sage_Logo_Synthesis_and_CVPR_2018_paper.pdf

 

7. Cross-View Image Synthesis Using Conditional GANs

（街区俯视图和直视转换）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Regmi_Cross-View_Image_Synthesis_CVPR_2018_paper.pdf

 

8. AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networks

（文本生成图片）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Xu_AttnGAN_Fine-Grained_Text_CVPR_2018_paper.pdf

 

9. High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

（图像高分辨率）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_High-Resolution_Image_Synthesis_CVPR_2018_paper.pdf

 

人脸相关

1. Finding Tiny Faces in the Wild with Generative Adversarial Network

（对低分辨率的人脸检测）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Bai_Finding_Tiny_Faces_CVPR_2018_paper.pdf

 

2. Learning Face Age Progression: A Pyramid Architecture of GANs

（预测年龄）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Yang_Learning_Face_Age_CVPR_2018_paper.pdf

 

3. Super-FAN: Integrated facial landmark localization and super-resolution of real-world low resolution faces in arbitrary poses with GANs

（对低分辨率人脸超分辨率）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Bulat_Super-FAN_Integrated_Facial_CVPR_2018_paper.pdf

4. Towards Open-Set Identity Preserving Face Synthesis

（人脸合成）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Bao_Towards_Open-Set_Identity_CVPR_2018_paper.pdf

 

5. Weakly Supervised Facial Action Unit Recognition through Adversarial Training

（人脸表情识别）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Peng_Weakly_Supervised_Facial_CVPR_2018_paper.pdf

 

6.FaceID-GAN: Learning a Symmetry Three-Player GAN for Identity-Preserving Face Synthesis

（生成多角度人脸）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Shen_FaceID-GAN_Learning_a_CVPR_2018_paper.pdf

 

7. UV-GAN: Adversarial Facial UV Map Completion for Pose-invariant Face Recognition

（人脸生成）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Deng_UV-GAN_Adversarial_Facial_CVPR_2018_paper.pdf

 

8.Face Aging with Identity-Preserved Conditional Generative Adversarial Networks

（人脸老化）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_Face_Aging_With_CVPR_2018_paper.pdf

 

人体相关

1. Deformable GANs for Pose-based Human Image Generation

（人物姿态迁移）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Siarohin_Deformable_GANs_for_CVPR_2018_paper.pdf

 

2. Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

(用GAN生成人行为轨迹追踪)

http://openaccess.thecvf.com/content_cvpr_2018/papers/Gupta_Social_GAN_Socially_CVPR_2018_paper.pdf

 

3. GANerated Hands for Real-Time 3D Hand Tracking from Monocular RGB

（用GAN生成的手势图片做手势追踪的数据集）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Mueller_GANerated_Hands_for_CVPR_2018_paper.pdf

 

4. Multistage Adversarial Losses for Pose-Based Human Image Synthesis

（人体姿态合成）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Si_Multistage_Adversarial_Losses_CVPR_2018_paper.pdf

5. Disentangled Person Image Generation

（人体合成）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Ma_Disentangled_Person_Image_CVPR_2018_paper.pdf

 

domain adaptation

（这个没来得及找了，可能转行咯~ 唉）

1. Generate to Adapt: Aligning Domains Using Generative Adversarial Networks

 

2. Re-Weighted Adversarial Adaptation Network for Unsupervised Domain Adaptation

 

3. Adversarial Feature Augmentation for Unsupervised Domain Adaptation

 

4. Domain Generalization With Adversarial Feature Learning

 

5. Image to Image Translation for Domain Adaptation

 

6. Duplex Generative Adversarial Network for Unsupervised Domain Adaptation

 

7. Conditional Generative Adversarial Network for Structured Domain Adaptation

 

目标跟踪检测

1.Generative Adversarial Learning Towards Fast Weakly Supervised Detection

（弱监督检测）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Shen_Generative_Adversarial_Learning_CVPR_2018_paper.pdf

 

2. SINT++: Robust Visual Tracking via Adversarial Positive Instance Generation

（对抗学习生成轨迹样本）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_SINT_Robust_Visual_CVPR_2018_paper.pdf

 

3. VITAL: VIsual Tracking via Adversarial Learning

http://openaccess.thecvf.com/content_cvpr_2018/papers/Song_VITAL_VIsual_Tracking_CVPR_2018_paper.pdf

 

GAN模型优化

1. SGAN: An Alternative Training of Generative Adversarial Network

（替代训练GAN）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Chavdarova_SGAN_An_Alternative_CVPR_2018_paper.pdf

 

2. GAGAN: Geometry-Aware Generative Adversarial Networks

（一种关注几何外形的GAN）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Kossaifi_GAGAN_Geometry-Aware_Generative_CVPR_2018_paper.pdf

 

3.Global versus Localized Generative Adversarial Nets

(局部优化GAN)

http://openaccess.thecvf.com/content_cvpr_2018/papers/Qi_Global_Versus_Localized_CVPR_2018_paper.pdf

 

4. Generative Adversarial Image Synthesis with Decision Tree Latent Controller

（决策树）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Kaneko_Generative_Adversarial_Image_CVPR_2018_paper.pdf

 

5. Unsupervised Deep Generative Adversarial Hashing Network

（哈希GAN）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Dizaji_Unsupervised_Deep_Generative_CVPR_2018_paper.pdf

 

6. Multi-Agent Diverse Generative Adversarial Networks

（多个生成器GAN）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Ghosh_Multi-Agent_Diverse_Generative_CVPR_2018_paper.pdf

 

7. Duplex Generative Adversarial Network for Unsupervised Domain Adaptation

（双鉴别器GAN）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Hu_Duplex_Generative_Adversarial_CVPR_2018_paper.pdf

 

图像分割

1. Translating and Segmenting Multimodal Medical Volumes With Cycle- and Shape-Consistency Generative Adversarial Network

（图像分割）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Zhang_Translating_and_Segmenting_CVPR_2018_paper.pdf

 

行人重识别

1. Person Transfer GAN to Bridge Domain Gap for Person Re-Identification

（用GAN生成的人体检测的图片）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Wei_Person_Transfer_GAN_CVPR_2018_paper.pdf

 

2. Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification

http://openaccess.thecvf.com/content_cvpr_2018/papers/Deng_Image-Image_Domain_Adaptation_CVPR_2018_paper.pdf

 

视觉特征提取

1. Visual Feature Attribution using Wasserstein GANs

http://openaccess.thecvf.com/content_cvpr_2018/papers/Baumgartner_Visual_Feature_Attribution_CVPR_2018_paper.pdf

 

域自适应学习

1. Generate To Adapt: Aligning Domains using Generative Adversarial Networks

（视觉域自适应）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Sankaranarayanan_Generate_to_Adapt_CVPR_2018_paper.pdf

 

图像检索

1. HashGAN: Deep Learning to Hash with Pair Conditional Wasserstein GAN

http://openaccess.thecvf.com/content_cvpr_2018/papers/Cao_HashGAN_Deep_Learning_CVPR_2018_paper.pdf

 

迁移学习

1.Partial Transfer Learning With Selective Adversarial Networks

http://openaccess.thecvf.com/content_cvpr_2018/papers/Cao_Partial_Transfer_Learning_CVPR_2018_paper.pdf

 

视频生成

1. MoCoGAN: Decomposing Motion and Content for Video Generation

（用GAN生成视频）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Tulyakov_MoCoGAN_Decomposing_Motion_CVPR_2018_paper.pdf

2. Learning to Generate Time-Lapse Videos Using Multi-Stage Dynamic Generative Adversarial Networks

（生成延时视频）

http://openaccess.thecvf.com/content_cvpr_2018/papers/Xiong_Learning_to_Generate_CVPR_2018_paper.pdf

 

小结：

  可以看出GAN相关的论文还不少呀，各个方面的都有，可是我个人觉得，可能没有那种特别厉害的吧~hh