Filters








14 Hits in 2.6 sec

StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generators [article]

Rinon Gal, Or Patashnik, Haggai Maron, Gal Chechik, Daniel Cohen-Or
<span title="2021-12-16">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Can a generative model be trained to produce images from a specific domain, guided by a text prompt only, without seeing any image? In other words: can an image generator be trained "blindly"?  ...  Leveraging the semantic power of large scale Contrastive-Language-Image-Pre-training (CLIP) models, we present a text-driven method that allows shifting a generative model to new domains, without having  ...  Conclusions We presented StyleGAN-NADA, a CLIP-guided zeroshot method for Non-Adversarial Domain Adaptation of image generators.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.00946v2">arXiv:2108.00946v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/lnn4ydsoenauxbpu6ijpm3ccn4">fatcat:lnn4ydsoenauxbpu6ijpm3ccn4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211224201607/https://arxiv.org/pdf/2108.00946v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/be/69/be697c79df8e4b280fec71751cb2d44667429f36.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.00946v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

One-Shot Adaptation of GAN in Just One CLIP [article]

Gihyun Kwon, Jong Chul Ye
<span title="2022-03-23">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
There are many recent research efforts to fine-tune a pre-trained generator with a few target images to generate images of a novel domain.  ...  Specifically, our model employs a two-step training strategy: reference image search in the source generator using a CLIP-guided latent optimization, followed by generator fine-tuning with a novel loss  ...  Especially StyleGAN-NADA further showed the versatility of model adaptation for a given target image. Few-shot domain adaptation.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.09301v2">arXiv:2203.09301v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/razjtcfruvc3ld3e47waxj6h2u">fatcat:razjtcfruvc3ld3e47waxj6h2u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220523222937/https://arxiv.org/pdf/2203.09301v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/0e/5e/0e5ed7f03582c0b7381af0f39aa14308fd25ec34.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.09301v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Mind the Gap: Domain Gap Control for Single Shot Domain Adaptation for Generative Adversarial Networks [article]

Peihao Zhu, Rameen Abdal, John Femiani, Peter Wonka
<span title="2021-11-28">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We present a new method for one shot domain adaptation. The input to our method is trained GAN that can produce images in domain A and a single reference image I_B from domain B.  ...  We propose several new regularizers for controlling the domain gap to optimize the weights of the pre-trained StyleGAN generator to output images in domain B instead of domain A.  ...  Together, these four loss terms guide the refinement process for G B . Among these losses, L clip across was proposed by StyleGAN-NADA (Gal et al., 2021) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.08398v2">arXiv:2110.08398v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ilfin2ybjrgxlhz3wblmyti6jq">fatcat:ilfin2ybjrgxlhz3wblmyti6jq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211022233644/https://arxiv.org/pdf/2110.08398v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/b2/6a/b26a3411076d3c6f0787f1dec3042e3075eab0dd.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.08398v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulation [article]

Gwanghyun Kim, Taesung Kwon, Jong Chul Ye
<span title="2022-04-05">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recently, GAN inversion methods combined with Contrastive Language-Image Pretraining (CLIP) enables zero-shot image manipulation guided by text prompts.  ...  Based on full inversion capability and high-quality image generation power of recent diffusion models, our method performs zero-shot image manipulation successfully even between unseen domains and takes  ...  VQGAN-CLIP [19, 41] recently show the impressive results of CLIP-guided conditional generation of artistic images.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.02711v4">arXiv:2110.02711v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zixp6naiobefvlv4gkkcmytgiu">fatcat:zixp6naiobefvlv4gkkcmytgiu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220407051114/https://arxiv.org/pdf/2110.02711v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/6a/4c/6a4c41277cc6b81aa43058258547d7cdd6a29163.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.02711v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Image-Based CLIP-Guided Essence Transfer [article]

Hila Chefer, Sagie Benaim, Roni Paiss, Lior Wolf
<span title="2022-03-17">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Through extensive experiments, we demonstrate the superiority of our methods for essence transfer over existing methods for style transfer, domain adaptation, and text-based semantic editing.  ...  Our blending operator combines the powerful StyleGAN generator and the semantic encoder of CLIP in a novel way that is simultaneously additive in both latent spaces, resulting in a mechanism that guarantees  ...  Our baselines include BlendGAN [47] and JoJoGAN [11] for face stylization, StyleGAN-NADA [13] and Mind The Gap (MTG) [23] for domain adaptation, and as a CLIP-aided text-based image editing method  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.12427v3">arXiv:2110.12427v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/z6fawah6knfmhlnvelrc2y2y34">fatcat:z6fawah6knfmhlnvelrc2y2y34</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220325184455/https://arxiv.org/pdf/2110.12427v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/d4/65/d4656e3b083f81cd2574f914907c82a2df3022b1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.12427v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

CtlGAN: Few-shot Artistic Portraits Generation with Contrastive Transfer Learning [article]

Yue Wang, Ran Yi, Ying Tai, Chengjie Wang, Lizhuang Ma
<span title="2022-03-16">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Existing portrait stylization models that generate good quality results are based on Image-to-Image Translation and require abundant data from both source and target domains.  ...  We adapt a pretrained StyleGAN in the source domain to a target artistic domain with no more than 10 artistic faces.  ...  StyleGAN-NADA [10] uses text to guide the domain adaptation by leveraging the semantic knowledge in the pretrained Contrastive-Language-Image-Pretraining (CLIP) [36] model.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.08612v1">arXiv:2203.08612v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/eqrot5yew5hd5iftuqofdqftpe">fatcat:eqrot5yew5hd5iftuqofdqftpe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220324170519/https://arxiv.org/pdf/2203.08612v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/62/f9/62f91259a04f935c8a44977c096eff16356267bb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.08612v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Multimodal Image Synthesis and Editing: A Survey [article]

Fangneng Zhan, Yingchen Yu, Rongliang Wu, Jiahui Zhang, Shijian Lu
<span title="2021-12-27">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We then describe multimodal image synthesis and editing approaches extensively with detailed frameworks including Generative Adversarial Networks (GANs), GAN Inversion, Transformers, and other methods  ...  We start with an introduction to different types of guidance modalities in image synthesis and editing.  ...  Bengio, “Neural machine trans- Or, “Stylegan-nada: Clip-guided domain adaptation of image lation by jointly learning to align and translate,” arXiv preprint generators  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.13592v1">arXiv:2112.13592v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hxkfyxbtbfgltju323os3xompe">fatcat:hxkfyxbtbfgltju323os3xompe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220103093453/https://arxiv.org/pdf/2112.13592v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/57/18/5718dce3400e8136d735c188bbb7520695dcc7b7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.13592v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

More Control for Free! Image Synthesis with Semantic Diffusion Guidance [article]

Xihui Liu, Dong Huk Park, Samaneh Azadi, Gong Zhang, Arman Chopikyan, Yuxiao Hu, Humphrey Shi, Anna Rohrbach, Trevor Darrell
<span title="2022-04-14">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We conduct experiments on FFHQ and LSUN datasets, and show results on fine-grained text-guided image synthesis, synthesis of images related to a style or content reference image, and examples with both  ...  We explore CLIP-based language guidance as well as both content and style-based image guidance in a unified framework.  ...  StyleCLIP [31] and StyleGAN-NADA [9] have demonstrated that CLIP enables text-guided image manipulation without domain-specific image-text pairs.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.05744v3">arXiv:2112.05744v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vlb23oi6ybh6hizkulhaqphuo4">fatcat:vlb23oi6ybh6hizkulhaqphuo4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211223160736/https://arxiv.org/pdf/2112.05744v2.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e3/57/e357f175516db266d066724139be05749926e530.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.05744v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Active Divergence with Generative Deep Learning – A Survey and Taxonomy [article]

Terence Broad, Sebastian Berns, Simon Colton, Mick Grierson
<span title="2021-07-12">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Generative deep learning systems offer powerful tools for artefact generation, given their ability to model distributions of data and generate high-fidelity results.  ...  use deep generative models in truly creative systems.  ...  The StyleGAN-NADA framework (Gal, 2021) takes advantage of the external knowledge of a contrastive language-image pre-training model (CLIP) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2107.05599v1">arXiv:2107.05599v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vfsapuewi5btvbmqe2ehtsvr5m">fatcat:vfsapuewi5btvbmqe2ehtsvr5m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210714230254/https://arxiv.org/pdf/2107.05599v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/09/1c/091c4ea2efaba23cd9024d8a063609c9a313b5cb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2107.05599v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Text2Mesh: Text-Driven Neural Stylization for Meshes [article]

Oscar Michel, Roi Bar-On, Richard Liu, Sagie Benaim, Rana Hanocka
<span title="2021-12-06">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In order to modify style, we obtain a similarity score between a text prompt (describing style) and a stylized mesh by harnessing the representational power of CLIP.  ...  Text2Mesh requires neither a pre-trained generative model nor a specialized 3D mesh dataset.  ...  Stylegan-nada: Clip-guided do- dataset, we are able to manipulate a myriad of meshes to main adaptation of image generators.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.03221v1">arXiv:2112.03221v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/2mfgjh37lna5hnjj6pvms6zuey">fatcat:2mfgjh37lna5hnjj6pvms6zuey</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211207134702/https://arxiv.org/pdf/2112.03221v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/d1/5b/d15b27edf3630728cdb40f49946365d9011641cf.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.03221v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

CLIP2StyleGAN: Unsupervised Extraction of StyleGAN Edit Directions [article]

Rameen Abdal, Peihao Zhu, John Femiani, Niloy J. Mitra, Peter Wonka
<span title="2021-12-09">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The success of StyleGAN has enabled unprecedented semantic editing capabilities, on both synthesized and real images.  ...  In another development, the CLIP architecture has been trained with internet-scale image and text pairings and has been shown to be useful in several zero-shot learning settings.  ...  Another CLIP based framework, StyleGAN-NADA [18] , uses the CLIP framework for zeroshot domain adaptation.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.05219v1">arXiv:2112.05219v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/o5ozhktixrgnpphjgy5gix5gmy">fatcat:o5ozhktixrgnpphjgy5gix5gmy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211214174906/https://arxiv.org/pdf/2112.05219v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/fc/7a/fc7aa4067fd44dd7c83790b9f7c1029c78ea775f.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.05219v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Semantic Segmentation In-the-Wild Without Seeing Any Segmentation Examples [article]

Nir Zabari, Yedid Hoshen
<span title="2021-12-06">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our method takes as input the image-level labels of the class categories present in the image; they can be obtained automatically or manually.  ...  We utilize a vision-language embedding model (specifically CLIP) to create a rough segmentation map for each class, using model interpretability methods.  ...  Stylegan-nada: Clip-guided domain adap- relevance maps can provide pseudo-supervision for exist- tation of image generators.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.03185v1">arXiv:2112.03185v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/k7tgvamso5frzkhqmxqrjs77am">fatcat:k7tgvamso5frzkhqmxqrjs77am</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211210092832/https://arxiv.org/pdf/2112.03185v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/d6/fb/d6fb5948be7b89d71620f25047c193a55d0d77d5.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.03185v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

State-of-the-Art in the Architecture, Methods and Applications of StyleGAN [article]

Amit H. Bermano and Rinon Gal and Yuval Alaluf and Ron Mokady and Yotam Nitzan and Omer Tov and Or Patashnik and Daniel Cohen-Or
<span title="2022-02-28">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Generative Adversarial Networks (GANs) have established themselves as a prevalent approach to image synthesis.  ...  However, the control offered by StyleGAN is inherently limited to the generator's learned distribution, and can only be applied to images generated by StyleGAN itself.  ...  Using the reference pair, in addition to the CLIP-space loss defined in StyleGAN-NADA, they introduce a new objective, which is to maintain CLIP-space direction similarity between each reference image  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.14020v1">arXiv:2202.14020v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qu3plbdnszdujcwxwq3zizysje">fatcat:qu3plbdnszdujcwxwq3zizysje</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220511165238/https://arxiv.org/pdf/2202.14020v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/36/71/36714983e9e70eaafe8d0f77d591e39e0c49eb0b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.14020v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

CLIPstyler: Image Style Transfer with a Single Text Condition [article]

Gihyun Kwon, Jong Chul Ye
<span title="2021-12-01">2021</span>
Using the pre-trained text-image embedding model of CLIP, we demonstrate the modulation of the style of content images only with a single text condition.  ...  Existing neural style transfer methods require reference style images to transfer texture information of style images to content images.  ...  For baseline models, we selected four different text-guided manipulation methods: StyleGAN-NADA [6] , StyleCLIP [21] , AdaIN+CLIP, VQGAN+CLIP.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.48550/arxiv.2112.00374">doi:10.48550/arxiv.2112.00374</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hlq4dl25trg2penxgs3v25lmra">fatcat:hlq4dl25trg2penxgs3v25lmra</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220223125635/https://arxiv.org/pdf/2112.00374.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/cc/ab/ccab419799395bb567f6cf6f7f501d27992e1ae3.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.48550/arxiv.2112.00374"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> Publisher / doi.org </button> </a>