Filters








10,599 Hits in 6.1 sec

Semantic photo manipulation with a generative image prior

David Bau, Hendrik Strobelt, William Peebles, Jonas Wulff, Bolei Zhou, Jun-Yan Zhu, Antonio Torralba
2019 ACM Transactions on Graphics  
Despite the recent success of GANs in synthesizing images conditioned on inputs such as a user sketch, text, or semantic labels, manipulating the high-level attributes of an existing natural photograph  ...  In this paper, we address these issues by adapting the image prior learned by GANs to image statistics of an individual image.  ...  METHOD Overview: We propose a general-purpose semantic photo manipulation method that integrates the natural image prior captured by a GAN generator. Figure 2 shows our image editing pipeline.  ... 
doi:10.1145/3306346.3323023 fatcat:ehpl5c4lxzf4lbrmun2ha4srpy

Generative Semantic Manipulation with Contrasting GAN [article]

Xiaodan Liang, Hao Zhang, Eric P. Xing
2017 arXiv   pre-print
Generative Adversarial Networks (GANs) have recently achieved significant improvement on paired/unpaired image-to-image translation, such as photo→ sketch and artist painting style transfer.  ...  Quantitative results further demonstrate the superiority of our model on generating manipulated results with high visual fidelity and reasonable object semantics.  ...  Compared to prior approaches that only transfer low-level information, we focus on high-level semantic manipulation on images given a desired category.  ... 
arXiv:1708.00315v1 fatcat:2ih6k4dukra4xkgflrvvhthc4a

Enjoy Your Editing: Controllable GANs for Image Editing via Latent Space Navigation [article]

Peiye Zhuang, Oluwasanmi Koyejo, Alexander G. Schwing
2021 arXiv   pre-print
Controllable semantic image editing enables a user to change entire image attributes with a few clicks, e.g., gradually making a summer scene look like it was taken in winter.  ...  Classic approaches for this task use a Generative Adversarial Net (GAN) to learn a latent space and suitable latent-space transformations.  ...  Semantic image editing seeks to automate image manipulation of semantics.  ... 
arXiv:2102.01187v3 fatcat:jsprls7hcjcv7bu6i33vszpbia

Search-based automatic image annotation via Flickr photos using tag expansion

Liang-Chi Hsieh, Winston H. Hsu
2010 2010 IEEE International Conference on Acoustics, Speech and Signal Processing  
Exponentially growing photo collections motivate the needs for automatic image annotation for effective manipulations (e.g., search, browsing).  ...  The intuition is to leverage surrounding tags from those visually similar Flickr photos for the unlabeled image. However, the tags are generally few and noisy.  ...  However, these methods are unreliable for Flickr photos generally with few and noisy usercontributed tags.  ... 
doi:10.1109/icassp.2010.5496215 fatcat:6ysc67uoejgx3i4bdosuuoztlu

Image Processing Using Multi-Code GAN Prior [article]

Jinjin Gu, Yujun Shen, Bolei Zhou
2020 arXiv   pre-print
The resulting high-fidelity image reconstruction enables the trained GAN models as prior to many real-world applications, such as image colorization, super-resolution, image inpainting, and semantic manipulation  ...  In this work, we propose a novel approach, called mGANprior, to incorporate the well-trained GANs as effective prior to a variety of image processing tasks.  ...  A recent work [3] applied generative image prior to semantic photo manipulation, but it can only edit some partial regions of the input image yet fails to apply to other tasks like colorization or super-resolution  ... 
arXiv:1912.07116v2 fatcat:cxb6ezakbre6hhktd6hmctpk2u

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs [article]

Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Andrew Tao, Jan Kautz, Bryan Catanzaro
2018 arXiv   pre-print
We present a new method for synthesizing high-resolution photo-realistic images from semantic label maps using conditional generative adversarial networks (conditional GANs).  ...  In this work, we generate 2048x1024 visually appealing results with a novel adversarial loss, as well as new multi-scale generator and discriminator architectures.  ...  JYZ is supported by a Facebook graduate fellowship.  ... 
arXiv:1711.11585v2 fatcat:ibvcptsa2fbejmf6cef5prab2q

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Andrew Tao, Jan Kautz, Bryan Catanzaro
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Abstract We present a new method for synthesizing highresolution photo-realistic images from semantic label maps using conditional generative adversarial networks (conditional GANs).  ...  synthesizing 2048 × 1024 images from semantic label maps (lower left corner in (a)).  ...  Acknowledgements We thank Taesung Park, Phillip Isola, Tinghui Zhou, Richard Zhang, Rafael Valle, and Alexei A. Efros for helpful comments.  ... 
doi:10.1109/cvpr.2018.00917 dblp:conf/cvpr/Wang0ZTKC18 fatcat:2klxsatsqnb3daa2edj7autp3i

DeepSEE: Deep Disentangled Semantic Explorative Extreme Super-Resolution [article]

Marcel C. Bühler, Andrés Romero, Radu Timofte
2020 arXiv   pre-print
In particular, it provides control of the semantic regions, their disentangled appearance and it allows a broad range of image manipulations.  ...  There are infinitely many plausible high-resolution variants for a given low-resolution natural image.  ...  Our model is trained with a strong low-resolution prior and hence, only allows relatively subtle shape manipulations.  ... 
arXiv:2004.04433v3 fatcat:you4taojxrdn7o5w6ywtjbqao4

Manipulating Attributes of Natural Scenes via Hallucination [article]

Levent Karacan, Zeynep Akata, Aykut Erdem, Erkut Erdem
2019 arXiv   pre-print
Once the scene is hallucinated with the given attributes, the corresponding look is then transferred to the input image while preserving the semantic details intact, giving a photo-realistic manipulation  ...  The key to our approach is a deep generative network which can hallucinate images of a scene as if they were taken at a different season (e.g. during winter), weather condition (e.g. in a cloudy day) or  ...  Once the scene is hallucinated with the given attributes, the corresponding look is then transferred to the input image while preserving the semantic details intact, giving a photo-realistic manipulation  ... 
arXiv:1808.07413v3 fatcat:74xfm7dieram5ex466qdjlzjmu

Face Sketch Synthesis via Semantic-Driven Generative Adversarial Network [article]

Xingqun Qi, Muyi Sun, Weining Wang, Xiaoxiao Dong, Qi Li, Caifeng Shan
2021 arXiv   pre-print
In addition, we exploit face parsing layouts as the semantic-level spatial prior to enforce globally structural style injection in the generator of SDGAN.  ...  Specifically, we conduct facial saliency detection on the input face photos to provide overall facial texture structure, which could be used as a global type of prior information.  ...  Recently, the image-to-image translation task has made great progress with the development of deep learning, especially with Generative Adversarial Network (GAN) [4] .  ... 
arXiv:2106.15121v1 fatcat:pouxuahtu5hbnnfja37xrv5dja

SemanticStyleGAN: Learning Compositional Generative Priors for Controllable Image Synthesis and Editing [article]

Yichun Shi, Xiao Yang, Yangyue Wan, Xiaohui Shen
2022 arXiv   pre-print
We present SemanticStyleGAN, where a generator is trained to model local semantic parts separately and synthesizes images in a compositional way.  ...  Thus, as a generic prior model with built-in disentanglement, it could facilitate the development of GAN-based applications and enable more potential downstream tasks.  ...  Controlled Synthesis and Image Editing With the semantic decomposition in the latent space, our model provides a more disentangled generative prior for image editing.  ... 
arXiv:2112.02236v3 fatcat:dvwd332znreenmgb6b4mipf6ia

ManiTrans: Entity-Level Text-Guided Image Manipulation via Token-wise Semantic Alignment and Generation [article]

Jianan Wang, Guansong Lu, Hang Xu, Zhenguo Li, Chunjing Xu, Yanwei Fu
2022 arXiv   pre-print
Our framework incorporates a semantic alignment module to locate the image regions to be manipulated, and a semantic loss to help align the relationship between the vision and language.  ...  Existing text-guided image manipulation methods aim to modify the appearance of the image or to edit a few objects in a virtual or simple scenario, which is far from practical application.  ...  Semantic Image Synthesis The task of semantic image synthesis aims to generate a photo-realistic image from a semantic label. Isola et al.  ... 
arXiv:2204.04428v1 fatcat:2c3i5fiucrbtroyfyqrceli4ri

Photo-realistic Facial Texture Transfer [article]

Parneet Kaur, Hang Zhang, Kristin J. Dana
2017 arXiv   pre-print
Our framework for face texture transfer (FaceTex) augments the prior work of MRF-CNN with a novel facial semantic regularization that incorporates a face prior regularization smoothly suppressing the changes  ...  We address the challenging problem of transferring face texture from a style face image to a content face image in a photorealistic manner without changing the identity of the original content image.  ...  A recent method [15] incorporates the Gram matrix with semantic segmentation and achieves high quality results for photo-realistic style transfer in scene images.  ... 
arXiv:1706.04306v1 fatcat:75ewb2dpizcddedmdtokqo75gm

Towards Open-World Text-Guided Face Image Generation and Manipulation [article]

Weihao Xia, Yujiu Yang, Jing-Hao Xue, Baoyuan Wu
2021 arXiv   pre-print
The latent codes can be randomly sampled from a prior distribution or inverted from a given image, which provides inherent supports for both image generation and manipulation from multi-modal inputs, such  ...  In this work, we propose a unified framework for both face image generation and manipulation that produces diverse and high-quality images with an unprecedented resolution at 1024 from multimodal inputs  ...  The goal is twofold: we want the generated images or manipulated attributes of given images to be firstly visually satisfactory and secondly semantically consistent with the given texts.  ... 
arXiv:2104.08910v1 fatcat:aexf7c26ancr7lwja5sonaj4ji

Semantic photo synthesis

Matthew Johnson, G. J. Brostow, J. Shotton, V. Kwatra, R. Cipolla, Bernice E. Rogowitz, Thrasyvoulos N. Pappas, Scott J. Daly
2007 Human Vision and Electronic Imaging XII  
automatically building an image library with semantic annotations from any photo collection.  ...  This paper presents a method that generates a composite image when a user types in nouns, such as "boat" and "sand."  ...  The result is an automatic means of generating a broad range of semantic labels for thousands of raw images.  ... 
doi:10.1117/12.720851 dblp:conf/hvei/0003BSKC07 fatcat:2kpc4b5f6banpkw5oyj7g6fnje
« Previous Showing results 1 — 15 out of 10,599 results