在最新的ICCV 2021会议中,涌现出大量基于生成对抗网络GAN的论文,广泛应用于各类视觉任务;本文在此做尽可能的梳理汇总!
下述论文已分类打包好!公众号后台回复 ICCV (长按红字、选中复制)获取分类、按文件夹汇总好的论文集,gan起来吧!!!
一、GAN改进
二、图像编辑-基于StyleGAN
三、图像编辑-逆映射
四、图像编辑-人脸
五、图像编辑-语义生成
六、图像恢复-超分
七、图像恢复-反光去除
八、图像恢复-光斑去除
九、图像恢复-去除阴影
十、图像恢复-水下图像失真去除
十一、图像恢复-修复
十二、图像检测-异常检测
十三、图像检测-拼接伪造
十四、图像检测-deepfake
十五、图像分割
十六、图像外插值-图像延展-外推
十七、图像转换
十八、文字生成图像
十九、说话人生成
二十、风格迁移
二十一、虚拟试衣
二十二、图像编辑-文本引导
二十三、图像编辑-单样本
二十四、开集识别
二十五、人脸识别
二十六、少样本生成
二十七、实例可控生成
二十八、视图合成
二十九、细粒度生成
三十、细粒度检索
三十一、解耦学习
三十二、可解释性
三十三、隐私保护
三十四、医学图像
一、GAN改进
1、Dual Contrastive Loss and Attention for GANs
2、Dual Projection Generative Adversarial Networks for Conditional Image Generation
3、Focal Frequency Loss for Image Reconstruction and Synthesis
4、Gradient Normalization for Generative Adversarial Networks
5、F-Drop&Match: GANs with a Dead Zone in the High-Frequency Domain
6、Latent Transformations via NeuralODEs for GAN-based Image Editing
7、Detail Me More: Improving GAN’s photo-realism of complex scenes
8、EigenGAN: Layer-Wise Eigen-Learning for GANs
9、Omni-GAN: On the Secrets of cGANs and Beyond
10、Towards Discovery and Attribution of Open-world GAN Generated Images
11、Unsupervised Image Generation with Infifinite Generative Adversarial Networks
二、图像编辑-基于StyleGAN
12、StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery
13、Diagonal Attention and Style-based GAN for Content-Style Disentanglement in Image Generation and Translation
14、ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinement
三、图像编辑-逆映射
15、From Continuity to Editability: Inverting GANs with Consecutive Images
16、GAN Inversion for Out-of-Range Images with Geometric Transformations
四、图像编辑-人脸
17、A Latent Transformer for Disentangled Face Editing in Images and Videos
五、图像编辑-语义生成
18、Collaging Class-specific GANs for Semantic Image Synthesis
19、Image Synthesis via Semantic Composition
20、Image Synthesis from Layout with Locality-Aware Mask Adaption
六、图像恢复-超分
21、Fourier Space Losses for Efficient Perceptual Image Super-Resolution
七、图像恢复-反光去除
22、Location-aware Single Image Reflection Removal
23、V-DESIRR: Very Fast Deep Embedded Single Image Reflection Removal
八、图像恢复-光斑去除
24、How to Train Neural Networks for Flare Removal
25、Light Source Guided Single-Image Flare Removal from Unpaired Data
九、图像恢复-去除阴影
26、DC-ShadowNet: Single-Image Hard and Soft Shadow Removal Using Unsupervised Domain-Classifier Guided Network
十、图像恢复-水下图像失真去除
27、Learning to Remove Refractive Distortions from Underwater Images
十一、图像恢复-修复
28 WaveFill: A Wavelet-based Generation Network for Image Inpainting
29 Painting from Part
30 High-Fidelity Pluralistic Image Completion with Transformers
31 Image Inpainting via Conditional Texture and Structure Dual Generation
32 Learning High-Fidelity Face Texture Completion without Complete Face Texture
33 Learning a Sketch Tensor Space for Image Inpainting of Man-made Scenes
34 Parallel Multi-Resolution Fusion Network for Image Inpainting
35 CR-Fill: Generative Image Inpainting with Auxiliary Contextual Reconstruction
十二、图像检测-异常检测
36 Learning Unsupervised Metaformer for Anomaly Detection
十三、图像检测-拼接伪造
37 Reality Transform Adversarial Generators for Image Splicing Forgery Detection and Localization
十四、图像检测-deepfake
38 Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Data
39 Joint Audio-Visual Deepfake Detection
40 Learning Self-Consistency for Deepfake Detection
十五、图像分割
41 Unsupervised Segmentation incorporating Shape Prior via Generative Adversarial Networks
42 InSeGAN: A Generative Approach to Segmenting Identical Instances in Depth Images
十六、图像外插值-图像延展-外推
43 SemIE: Semantically-aware Image Extrapolation
十七、图像转换
44、 SPatchGAN: A Statistical Feature Based Discriminator for Unsupervised Image-to-Image Translation
45、 Dual Transfer Learning for Event-based End-task Prediction via Pluggable Event to Image Translation
46、 Bridging the Gap between Label- and Reference-based Synthesis in Multi-attribute Image-to-Image Translation
47、 Frequency Domain Image Translation: More Photo-realistic, Better Identity-preserving
48、 Harnessing the Conditioning Sensorium for Improved Image Translation
49、 Instance-wise Hard Negative Example Generation for Contrastive Learning in Unpaired Image-to-Image Translation
59、 Online Multi-Granularity Distillation for GAN Compression
51、 Rethinking the Truly Unsupervised Image-to-Image Translation
52、 Scaling-up Disentanglement for Image Translation
53、 Semantically Robust Unpaired Image Translation for Data with Unmatched Semantics Statistics
54、 TransferI2I: Transfer Learning for Image-to-Image Translation from Small Datasets
55、Unaligned Image-to-Image Translation by Learning to Reweight
56、Sketch Your Own GAN
十八、文字生成图像
57、 DAE-GAN: Dynamic Aspect-aware GAN for Text-to-Image Synthesis
十九、说话人生成
58、 Talk-to-Edit: Fine-Grained Facial Editing via Dialog
59、 HeadGAN: One-shot Neural Head Synthesis and Editing
60、Learned Spatial Representations for Few-shot Talking-Head Synthesis
61、 MeshTalk: 3D Face Animation from Speech using Cross-Modality Disentanglement
62、 FACIAL: Synthesizing Dynamic Talking Face with Implicit Attribute Learning
二十、风格迁移
63、 DRB-GAN: A Dynamic ResBlock Generative Adversarial Network for Artistic Style Transfer
64、 Diverse Image Style Transfer via Invertible Cross-Space Mapping
二十一、虚拟试衣
65、FashionMirror: Co-attention Feature-remapping Virtual Try-on with Sequential Template Poses
66、Dressing in Order: Recurrent Person Image Generation for Pose Transfer, Virtual Try-on and Outfit Editing
67、Structure-transformed Texture-enhanced Network for Person Image Synthesis
二十二、图像编辑-文本引导
68、Language-Guided Global Image Editing via Cross-Modal Cyclic Mechanism
二十三、图像编辑-单样本
69、Image Shape Manipulation from a Single Augmented Training Sample
二十四、开集识别
70、OpenGAN: Open-Set Recognition via Open Data Generation
二十五、人脸识别
71、SynFace: Face Recognition with Synthetic Data
72、Learning Facial Representations from the Cycle-consistency of Face
73、Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition
二十六、少样本生成
74、LoFGAN: Fusing Local Representations for Few-shot Image Generation
二十七、实例可控生成
75、Multi-Class Multi-Instance Count Conditioned Adversarial Image Generation
二十八、视图合成
76、Infifinite Nature: Perpetual View Generation of Natural Scenes from a Single Image
二十九、细粒度生成
77、Semi-Supervised Single-Stage Controllable GANs for Conditional Fine-Grained Image Generation
三十、细粒度检索
78、Retrieve in Style: Unsupervised Facial Feature Transfer and Retrieval
79、Face Image Retrieval with Attribute Manipulation
三十一、解耦学习
80、GAN-Control: Explicitly Controllable GANs
81、Video Autoencoder: self-supervised disentanglement of static 3D structure and motion
三十二、可解释性
82、WarpedGANSpace: Finding non-linear RBF paths in GAN latent space
83、Toward a Visual Concept Vocabulary for GAN Latent Space
84、When do GANs replicate? On the choice of dataset size
85、Explaining in Style: Training a GAN to explain a classifier in StyleSpace
三十三、隐私保护
86、Personalized and Invertible Face De-identification by Disentangled Identity Information Manipulation
87、Towards Face Encryption by Generating Adversarial Identity Masks
三十四、医学图像
88、Mutual-Complementing Framework for Nuclei Detection and Segmentation in Pathology Image
89、Long-Term Temporally Consistent Unpaired Video Translation from Simulated Surgical 3D Data
90、Collaborative and Adversarial Learning of Focused and Dispersive Representations for Semi-supervised Polyp Segmentation
猜您喜欢:
ICCV 2021 | 35个GAN主题,最全GAN论文汇总
附下载 |《TensorFlow 2.0 深度学习算法实战》