Image harmonization aims to produce visually harmonious composite images by adjusting the foreground appearance to be compatible with the background. When the composite image has photographic foreground and painterly background, the task is called painterly image harmonization. There are only few works on this task, which are either time-consuming or weak in generating well-harmonized results. In this work, we propose a novel painterly harmonization network consisting of a dual-domain generator and a dual-domain discriminator, which harmonizes the composite image in both spatial domain and frequency domain. The dual-domain generator performs harmonization by using AdaIn modules in the spatial domain and our proposed ResFFT modules in the frequency domain. The dual-domain discriminator attempts to distinguish the inharmonious patches based on the spatial feature and frequency feature of each patch, which can enhance the ability of generator in an adversarial manner. Extensive experiments on the benchmark dataset show the effectiveness of our method. Our code and model are available at https://github.com/bcmi/PHDNet-Painterly-Image-Harmonization.
translated by 谷歌翻译
图像构成目标在将前景对象插入到背景图像中。最先前的图像构成方法专注于调整前景,使其与背景兼容,同时忽略背景的前景的阴影效果。在这项工作中,我们专注于为复合图像中的前景对象产生合理的阴影。首先,我们通过基于配对的真实图像和deshadowed图像生成合成合成图像来贡献实际阴影生成数据集脱差。然后,我们提出了一种新的阴影生成网络SGRNet,其包括阴影掩模预测阶段和阴影填充阶段。在阴影掩模预测阶段,前景和背景信息彻底互动以产生前景影掩模。在阴影填充阶段,预计暗影参数填充阴影区域。我们的Desoba数据集和真实复合图像的广泛实验证明了我们所提出的方法的有效性。我们的数据集和代码可在https://github.com/bcmi/object-shadow-generation-dataset-desoba获得。
translated by 谷歌翻译
作为一个常见的图像编辑操作,图像组成旨在将前景从一个图像切割并粘贴在另一个图像上,从而产生复合图像。但是,有许多问题可能使复合图像不现实。这些问题可以总结为前景和背景之间的不一致,包括外观不一致(例如,不兼容的照明),几何不一致(例如不合理的大小)和语义不一致(例如,不匹配的语义上下文)。先前的作品将图像组成任务分为多个子任务,其中每个子任务在一个或多个问题上目标。具体而言,对象放置旨在为前景找到合理的比例,位置和形状。图像混合旨在解决前景和背景之间的不自然边界。图像协调旨在调整前景的照明统计数据。影子生成旨在为前景产生合理的阴影。通过将所有上述努力放在一起,我们可以获取现实的复合图像。据我们所知,以前没有关于图像组成的调查。在本文中,我们对图像组成的子任务进行了全面的调查。对于每个子任务,我们总结了传统方法,基于深度学习的方法,数据集和评估。我们还指出了每个子任务中现有方法的局限性以及整个图像组成任务的问题。图像组合的数据集和代码在https://github.com/bcmi/awesome-image-composition上进行了总结。
translated by 谷歌翻译
任意样式转移生成了艺术图像,该图像仅使用一个训练有素的网络结合了内容图像的结构和艺术风格的结合。此方法中使用的图像表示包含内容结构表示和样式模式表示形式,这通常是预训练的分类网络中高级表示的特征表示。但是,传统的分类网络是为分类而设计的,该分类通常集中在高级功能上并忽略其他功能。结果,风格化的图像在整个图像中均匀地分布了样式元素,并使整体图像结构无法识别。为了解决这个问题,我们通过结合全球和局部损失,引入了一种新型的任意风格转移方法,并通过结构增强。局部结构细节由LapStyle表示,全局结构由图像深度控制。实验结果表明,与其他最新方法相比,我们的方法可以在几个常见数据集中生成具有令人印象深刻的视觉效果的更高质量图像。
translated by 谷歌翻译
Image harmonization task aims at harmonizing different composite foreground regions according to specific background image. Previous methods would rather focus on improving the reconstruction ability of the generator by some internal enhancements such as attention, adaptive normalization and light adjustment, $etc.$. However, they pay less attention to discriminating the foreground and background appearance features within a restricted generator, which becomes a new challenge in image harmonization task. In this paper, we propose a novel image harmonization framework with external style fusion and region-wise contrastive learning scheme. For the external style fusion, we leverage the external background appearance from the encoder as the style reference to generate harmonized foreground in the decoder. This approach enhances the harmonization ability of the decoder by external background guidance. Moreover, for the contrastive learning scheme, we design a region-wise contrastive loss function for image harmonization task. Specifically, we first introduce a straight-forward samples generation method that selects negative samples from the output harmonized foreground region and selects positive samples from the ground-truth background region. Our method attempts to bring together corresponding positive and negative samples by maximizing the mutual information between the foreground and background styles, which desirably makes our harmonization network more robust to discriminate the foreground and background style features when harmonizing composite images. Extensive experiments on the benchmark datasets show that our method can achieve a clear improvement in harmonization quality and demonstrate the good generalization capability in real-scenario applications.
translated by 谷歌翻译
STYLE TRANSED引起了大量的关注,因为它可以在保留图像结构的同时将给定图像更改为一个壮观的艺术风格。然而,常规方法容易丢失图像细节,并且在风格转移期间倾向于产生令人不快的伪影。在本文中,为了解决这些问题,提出了一种具有目标特征调色板的新颖艺术程式化方法,可以准确地传递关键特征。具体而言,我们的方法包含两个模块,即特征调色板组成(FPC)和注意着色(AC)模块。 FPC模块基于K-means群集捕获代表特征,并生成特征目标调色板。以下AC模块计算内容和样式图像之间的注意力映射,并根据注意力映射和目标调色板传输颜色和模式。这些模块使提出的程式化能够专注于关键功能并生成合理的传输图像。因此,所提出的方法的贡献是提出一种新的深度学习的样式转移方法和当前目标特征调色板和注意着色模块,并通过详尽的消融研究提供对所提出的方法的深入分析和洞察。定性和定量结果表明,我们的程式化图像具有最先进的性能,具有保护核心结构和内容图像的细节。
translated by 谷歌翻译
当使用切割和贴子获取复合图像时,前景和背景之间的几何不一致可能会严重损害其保真度。为了解决复合图像中的几何不一致,几项现有作品学会了扭曲前景对象的几何校正。但是,没有注释的数据集导致性能不令人满意和不可靠的评估。在这项工作中,我们为涵盖三个典型应用程序方案的虚拟尝试(Strat)数据集做出了空间转换。此外,以前的作品仅将前景和背景作为输入,而无需考虑它们的相互对应。取而代之的是,我们提出了一个新颖的对应学习网络(CORRELNET),以使用交叉注意图对正面和背景之间的对应关系进行建模,我们可以预测目标坐标,即前景的每个源坐标都应映射到背景上。然后,前景对象的翘曲参数可以从源和目标坐标对派生。此外,我们学习一个过滤面膜,以消除嘈杂的坐标对,以估计更准确的翘曲参数。我们的Strat数据集上的广泛实验表明,我们所提出的CORRELNET对以前的方法更有利。
translated by 谷歌翻译
本文的目标是对面部素描合成(FSS)问题进行全面的研究。然而,由于获得了手绘草图数据集的高成本,因此缺乏完整的基准,用于评估过去十年的FSS算法的开发。因此,我们首先向FSS引入高质量的数据集,名为FS2K,其中包括2,104个图像素描对,跨越三种类型的草图样式,图像背景,照明条件,肤色和面部属性。 FS2K与以前的FSS数据集不同于难度,多样性和可扩展性,因此应促进FSS研究的进展。其次,我们通过调查139种古典方法,包括34个手工特征的面部素描合成方法,37个一般的神经式传输方法,43个深映像到图像翻译方法,以及35个图像 - 素描方法。此外,我们详细说明了现有的19个尖端模型的综合实验。第三,我们为FSS提供了一个简单的基准,名为FSGAN。只有两个直截了当的组件,即面部感知屏蔽和风格矢量扩展,FSGAN将超越所提出的FS2K数据集的所有先前最先进模型的性能,通过大边距。最后,我们在过去几年中汲取的经验教训,并指出了几个未解决的挑战。我们的开源代码可在https://github.com/dengpingfan/fsgan中获得。
translated by 谷歌翻译
最近的研究表明,通用风格转移的成功取得了巨大的成功,将任意视觉样式转移到内容图像中。但是,现有的方法遭受了审美的非现实主义问题,该问题引入了不和谐的模式和明显的人工制品,从而使结果很容易从真实的绘画中发现。为了解决这一限制,我们提出了一种新颖的美学增强风格转移方法,可以在美学上为任意风格产生更现实和令人愉悦的结果。具体而言,我们的方法引入了一种审美歧视者,以从大量的艺术家创造的绘画中学习通用的人类自愿美学特征。然后,合并了美学特征,以通过新颖的美学感知样式(AESSA)模块来增强样式转移过程。这样的AESSA模块使我们的Aesust能够根据样式图像的全局美学通道分布和内容图像的局部语义空间分布有效而灵活地集成样式模式。此外,我们还开发了一种新的两阶段转移培训策略,并通过两种审美正规化来更有效地训练我们的模型,从而进一步改善风格化的性能。广泛的实验和用户研究表明,我们的方法比艺术的状态综合了美学上更加和谐和现实的结果,从而大大缩小了真正的艺术家创造的绘画的差异。我们的代码可在https://github.com/endywon/aesust上找到。
translated by 谷歌翻译
对象放置旨在将前景对象放在具有合适位置和大小的背景图像上。在这项工作中,我们将对象放置视为图形完成问题,并提出一个新的图形完成模块(GCM)。背景场景由一个图形表示,在不同的空间位置具有多个节点,并带有各种接收场。前景对象被编码为应插入该图中合理位置的特殊节点。我们还在GCM的结构上设计了一个双路径框架,以完全利用带注释的复合图像。通过在OPA数据集上进行广泛的实验,我们的方法证明在生成合理的对象放置而不会丧失多样性方面显着胜过现有的方法。
translated by 谷歌翻译
Gatys et al. recently introduced a neural algorithm that renders a content image in the style of another image, achieving so-called style transfer. However, their framework requires a slow iterative optimization process, which limits its practical application. Fast approximations with feed-forward neural networks have been proposed to speed up neural style transfer. Unfortunately, the speed improvement comes at a cost: the network is usually tied to a fixed set of styles and cannot adapt to arbitrary new styles. In this paper, we present a simple yet effective approach that for the first time enables arbitrary style transfer in real-time. At the heart of our method is a novel adaptive instance normalization (AdaIN) layer that aligns the mean and variance of the content features with those of the style features. Our method achieves speed comparable to the fastest existing approach, without the restriction to a pre-defined set of styles. In addition, our approach allows flexible user controls such as content-style trade-off, style interpolation, color & spatial controls, all using a single feed-forward neural network.
translated by 谷歌翻译
面部去夹旨在从模糊的输入图像恢复清晰的面部图像,具有更明确的结构和面部细节。然而,大多数传统的图像和面部去夹方法的重点是整个产生的图像分辨率,而不考虑特殊的面部纹理并且通常产生无充气的细节。考虑到面部和背景具有不同的分布信息,在本研究中,我们设计了一种基于可分离的归一化和自适应非规范化(SnAdnet)的有效面部去孔网络。首先,我们微调面部解析网络以获得精确的面部结构。然后,我们将脸部解析功能划分为面部前景和背景。此外,我们构建了一种新的特征自适应非规范化,以将FAYCIAL结构规则为辅助的条件,以产生更加和谐的面部结构。另外,我们提出了一种纹理提取器和多贴片鉴别器,以增强所生成的面部纹理信息。 Celeba和Celeba-HQ数据集的实验结果表明,所提出的面部去孔网络以更具面部细节恢复面部结构,并在结构相似性索引方法(SSIM),峰值信号方面对最先进的方法进行有利的方法。信噪比(PSNR),Frechet Inception距离(FID)和L1以及定性比较。
translated by 谷歌翻译
我们提出了一个极其简单的超分辨率样式转移框架,称为URST,以灵活地处理任意的高分辨率图像(例如,10000x10000像素)第一次转移。由于在处理超高分辨率图像时,由于巨大的内存成本和小行程大小,大多数现有最先进的方法将降低。 URST完全避免了由超高分辨率图像引起的内存问题(1)将图像划分为小块和(2)与新颖的缩略图实例归一化(TIN)执行修补程序样式传输。具体而言,TIN可以提取缩略图功能的归一化统计信息,并将它们应用于小补丁,确保不同补丁之间的风格一致性。总的来说,与现有技术相比,URST框架有三个优点。 (1)我们将输入图像分为小补丁并采用锡,成功传输图像样式,具有任意的高分辨率。 (2)实验表明,我们的URST超越了现有的SOTA方法对超高分辨率图像,从提高行程大小的提出的中风感知损失的有效性中受益。 (3)我们的URST可以轻松插入大多数现有的样式转移方法,即使在没有培训的情况下也直接提高他们的性能。代码可在https://git.io/urst上获得。
translated by 谷歌翻译
Automatic font generation without human experts is a practical and significant problem, especially for some languages that consist of a large number of characters. Existing methods for font generation are often in supervised learning. They require a large number of paired data, which are labor-intensive and expensive to collect. In contrast, common unsupervised image-to-image translation methods are not applicable to font generation, as they often define style as the set of textures and colors. In this work, we propose a robust deformable generative network for unsupervised font generation (abbreviated as DGFont++). We introduce a feature deformation skip connection (FDSC) to learn local patterns and geometric transformations between fonts. The FDSC predicts pairs of displacement maps and employs the predicted maps to apply deformable convolution to the low-level content feature maps. The outputs of FDSC are fed into a mixer to generate final results. Moreover, we introduce contrastive self-supervised learning to learn a robust style representation for fonts by understanding the similarity and dissimilarities of fonts. To distinguish different styles, we train our model with a multi-task discriminator, which ensures that each style can be discriminated independently. In addition to adversarial loss, another two reconstruction losses are adopted to constrain the domain-invariant characteristics between generated images and content images. Taking advantage of FDSC and the adopted loss functions, our model is able to maintain spatial information and generates high-quality character images in an unsupervised manner. Experiments demonstrate that our model is able to generate character images of higher quality than state-of-the-art methods.
translated by 谷歌翻译
Photo-realistic style transfer aims at migrating the artistic style from an exemplar style image to a content image, producing a result image without spatial distortions or unrealistic artifacts. Impressive results have been achieved by recent deep models. However, deep neural network based methods are too expensive to run in real-time. Meanwhile, bilateral grid based methods are much faster but still contain artifacts like overexposure. In this work, we propose the \textbf{Adaptive ColorMLP (AdaCM)}, an effective and efficient framework for universal photo-realistic style transfer. First, we find the complex non-linear color mapping between input and target domain can be efficiently modeled by a small multi-layer perceptron (ColorMLP) model. Then, in \textbf{AdaCM}, we adopt a CNN encoder to adaptively predict all parameters for the ColorMLP conditioned on each input content and style image pair. Experimental results demonstrate that AdaCM can generate vivid and high-quality stylization results. Meanwhile, our AdaCM is ultrafast and can process a 4K resolution image in 6ms on one V100 GPU.
translated by 谷歌翻译
提供和渲染室内场景一直是室内设计的一项长期任务,艺术家为空间创建概念设计,建立3D模型的空间,装饰,然后执行渲染。尽管任务很重要,但它很乏味,需要巨大的努力。在本文中,我们引入了一个特定领域的室内场景图像合成的新问题,即神经场景装饰。鉴于一张空的室内空间的照片以及用户确定的布局列表,我们旨在合成具有所需的家具和装饰的相同空间的新图像。神经场景装饰可用于以简单而有效的方式创建概念室内设计。我们解决这个研究问题的尝试是一种新颖的场景生成体系结构,它将空的场景和对象布局转化为现实的场景照片。我们通过将其与有条件图像合成基线进行比较,以定性和定量的方式将其进行比较,证明了我们提出的方法的性能。我们进行广泛的实验,以进一步验证我们生成的场景的合理性和美学。我们的实现可在\ url {https://github.com/hkust-vgd/neural_scene_decoration}获得。
translated by 谷歌翻译
图像协调旨在根据具体背景修改复合区域的颜色。以前的工作模型是使用Unet系列结构的像素-ID映像转换。然而,模型大小和计算成本限制了模型在边缘设备和更高分辨率图像上的能力。为此,我们首次提出了一种新的空间分离曲线渲染网络(S $ ^ 2 $ CRNET),首次进行高效和高分辨率的图像协调。在S $ ^ 2 $ CRNET中,我们首先将屏蔽前景和背景的缩略图中提取空间分离的嵌入物。然后,我们设计一种曲线渲染模块(CRM),其使用线性层学习并结合空间特定知识,以生成前景区域中的方向曲线映射的参数。最后,我们使用学习的颜色曲线直接渲染原始的高分辨率图像。此外,我们还通过Cascaded-CRM和语义CRM分别进行了两个框架的延伸,分别用于级联细化和语义指导。实验表明,与以前的方法相比,该方法降低了90%以上的参数,但仍然达到了合成的iHarmony4和现实世界DIH测试集的最先进的性能。此外,我们的方法可以在0.1秒内在更高分辨率图像(例如,2048美元\ times2048 $)上顺利工作,而不是所有现有方法的GPU计算资源。代码将在\ url {http://github.com/stefanleong/s2crnet}中提供。
translated by 谷歌翻译
在本文中,我们介绍了纹理改革器,一个快速和通用的神经基础框架,用于使用用户指定的指导进行交互式纹理传输。挑战在三个方面:1)任务的多样性,2)引导图的简单性,以及3)执行效率。为了解决这些挑战,我们的主要思想是使用由i)全球视图结构对准阶段,ii)局部视图纹理细化阶段和III)的新的前馈多视图和多级合成程序。效果增强阶段用相干结构合成高质量结果,并以粗略的方式进行细纹细节。此外,我们还介绍了一种新颖的无学习视图特定的纹理改革(VSTR)操作,具有新的语义地图指导策略,以实现更准确的语义引导和结构保存的纹理传输。关于各种应用场景的实验结果展示了我们框架的有效性和优越性。并与最先进的交互式纹理转移算法相比,它不仅可以实现更高的质量结果,而且更加显着,也是更快的2-5个数量级。代码可在https://github.com/endywon/texture --reformer中找到。
translated by 谷歌翻译
Arbitrary style transfer (AST) transfers arbitrary artistic styles onto content images. Despite the recent rapid progress, existing AST methods are either incapable or too slow to run at ultra-resolutions (e.g., 4K) with limited resources, which heavily hinders their further applications. In this paper, we tackle this dilemma by learning a straightforward and lightweight model, dubbed MicroAST. The key insight is to completely abandon the use of cumbersome pre-trained Deep Convolutional Neural Networks (e.g., VGG) at inference. Instead, we design two micro encoders (content and style encoders) and one micro decoder for style transfer. The content encoder aims at extracting the main structure of the content image. The style encoder, coupled with a modulator, encodes the style image into learnable dual-modulation signals that modulate both intermediate features and convolutional filters of the decoder, thus injecting more sophisticated and flexible style signals to guide the stylizations. In addition, to boost the ability of the style encoder to extract more distinct and representative style signals, we also introduce a new style signal contrastive loss in our model. Compared to the state of the art, our MicroAST not only produces visually superior results but also is 5-73 times smaller and 6-18 times faster, for the first time enabling super-fast (about 0.5 seconds) AST at 4K ultra-resolutions. Code is available at https://github.com/EndyWon/MicroAST.
translated by 谷歌翻译
Arbitrary Style Transfer is a technique used to produce a new image from two images: a content image, and a style image. The newly produced image is unseen and is generated from the algorithm itself. Balancing the structure and style components has been the major challenge that other state-of-the-art algorithms have tried to solve. Despite all the efforts, it's still a major challenge to apply the artistic style that was originally created on top of the structure of the content image while maintaining consistency. In this work, we solved these problems by using a Deep Learning approach using Convolutional Neural Networks. Our implementation will first extract foreground from the background using the pre-trained Detectron 2 model from the content image, and then apply the Arbitrary Style Transfer technique that is used in SANet. Once we have the two styled images, we will stitch the two chunks of images after the process of style transfer for the complete end piece.
translated by 谷歌翻译