Semantic portrait color transfer with internet images

Yang, Yue; Zhao, Hanli; You, Lihua; Tu, Renlong; Wu, Xueyi; Jin, Xiaogang

doi:10.1007/s11042-015-3063-x

Cited by 36 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method can also be utilized to understand the optimal solutions to other optimization problems, such as regression tasks [163], covert communication system [164]- [166], shape registration [167], micro-expression spotting [168], [169], medical diagnosis [170]- [173], image editing [174]- [176], engineering optimization problems [177], [178], brain function prediction [179], [180], service ecosystem [181], [182], image dehazing [183]- [185], epidemic prevention and control [186], [187], large scale network analysis [188], energy storage planning and scheduling [189], Lunar impact crater identification and age estimation [190], social recommendation and QOS-aware service composition [191]- [193], 3D deformable shape analysis [194], [195], and feature selection [196]- [198].…”

Section: Discussionmentioning

confidence: 99%

Early Recognition and Discrimination of COVID-19 Severity Using Slime Mould Support Vector Machine for Medical Decision-Making

et al. 2021

View full text Add to dashboard Cite

has spread rapidly across the world, leading to the insufficiency of medical resources in many regions. Early detection and identification of high-risk COVID-19 patients will contribute to early intervention and optimize medical resource allocation. Using the clinical data from the Affiliated Yueqing Hospital of Wenzhou Medical University (Yueqing, China), an evolutionary support vector machine model is designed to recognize and discriminate the severity of the COVID-19 by patients basic information and hematological indexes. The support vector machine is a frequently used pattern classification tool affected by both the kernel parameter setting and feature selection for its classification accuracy. This study recommends an enhanced Slime Mould Algorithm (ESMA), mixing a new movement strategy of white holes, black holes, and wormholes, to perform parameter optimization and feature selection simultaneously for SVM. Therefore, the proposed SVM framework (ESMA-SVM) can also obtain highquality classification results, and it is less prone to stagnation in the classification process. To verify the capabilities of the proposed methodology, first, the performance of the ESMA is thoroughly verified by using IEEE CEC2017 benchmark functions and the diversity and compared with other similar methods experimentally using these standard benchmark functions. Moreover, the balance between diversification and intensification capability of the enhanced ESMA and the original SMA is also investigated statistically. Finally, the designed model ESMA-SVM and other competitive SVM models based on other optimization algorithms are applied to early recognition and discrimination of COVID-19 severity. Through the analysis of experimental results, the core compensations of ESMA are confirmed, and the ESMA-SVM can obtain strong performance in terms of several performance evaluation indexes on discrimination of COVID-19 severity.

show abstract

Section: Discussionmentioning

confidence: 99%

Early Recognition and Discrimination of COVID-19 Severity Using Slime Mould Support Vector Machine for Medical Decision-Making

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Chia et al [5] require a pre-segmented target image with semantic labels and search within Internet images with associated semantic labels to colorize the target image. The method is also generalized to semantic portrait color transfer [28]. Deshpande et al [29] propose an automated method for image colorization that learns colorization from a set of examples (rather than one reference image) by exploiting the LEARCH (learning-tosearch) framework.…”

Section: Related Workmentioning

confidence: 99%

Example-Based Image Colorization Using Locality Consistent Sparse Representation

Zhao

et al. 2017

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Image colorization aims to produce a natural looking color image from a given gray-scale image, which remains a challenging problem. In this paper, we propose a novel example-based image colorization method exploiting a new locality consistent sparse representation. Given a single reference color image, our method automatically colorizes the target gray-scale image by sparse pursuit. For efficiency and robustness, our method operates at the superpixel level. We extract low-level intensity features, mid-level texture features, and high-level semantic features for each superpixel, which are then concatenated to form its descriptor. The collection of feature vectors for all the superpixels from the reference image composes the dictionary. We formulate colorization of target superpixels as a dictionary-based sparse reconstruction problem. Inspired by the observation that superpixels with similar spatial location and/or feature representation are likely to match spatially close regions from the reference image, we further introduce a locality promoting regularization term into the energy formulation, which substantially improves the matching consistency and subsequent colorization results. Target superpixels are colorized based on the chrominance information from the dominant reference superpixels. Finally, to further improve coherence while preserving sharpness, we develop a new edge-preserving filter for chrominance channels with the guidance from the target gray-scale image. To the best of our knowledge, this is the first work on sparse pursuit image colorization from single reference images. Experimental results demonstrate that our colorization method outperforms the state-of-the-art methods, both visually and quantitatively using a user study.

show abstract

“…The majority of techniques equalizes the mean and variance of a style and content image to control color distributions via luminance-based (Reinhard et al 2001) or HSL-based (Neumann and Neumann 2005) histograms. Extensions integrate feature maps to consider local information as well, such as image segmentation (Wu et al 2013;Xiao and Ma 2009), edge-aware texture descriptors (Arbelot et al 2016), and semantics (Yang et al 2017) to colorize grayscale images. With interactive methods it is also possible to maintain control over a set of colors that is involved in palette-based color transfers (Chang et al 2015;Pouli and Reinhard 2011).…”

Section: Style Transfer Using Image Statisticsmentioning

confidence: 99%

Neural style transfer

Semmo

Isenberg

Döllner

2017

Proceedings of the Symposium on Non-Photorealistic Animation and Rendering

View full text Add to dashboard Cite

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

Semantic portrait color transfer with internet images

Cited by 36 publications

References 24 publications

Early Recognition and Discrimination of COVID-19 Severity Using Slime Mould Support Vector Machine for Medical Decision-Making

Early Recognition and Discrimination of COVID-19 Severity Using Slime Mould Support Vector Machine for Medical Decision-Making

Example-Based Image Colorization Using Locality Consistent Sparse Representation

Neural style transfer

Contact Info

Product

Resources

About