Academic literature on the topic 'Photographie colorisée'

L’analyse des articles publiés dans la seconde moitié du xixe siècle au sujet de la coloration à la main des photographies montre que, malgré les tentatives initiales de faire de cette pratique une profession « masculine » exigeant savoir-faire et talent artistique, le rejet de la coloration par l’establishment, ainsi que les efforts des propriétaires de studios pour produire en série des images coloriées par une main-d’œuvre payée à la tâche, ont relégué les coloristes à un rang inférieur. Mal considérée et rémunérée, cette activité était ouverte aux femmes d’Europe et d’Amérique, à une époque où d’autres emplois au sein des studios photographiques leur étaient souvent inaccessibles. Toutefois, ces coloristes dotées d’une vaste gamme de compétences se sont volontiers consacrées à cette pratique nouvelle, que leur recommandaient toutes sortes de porte-parole dans la presse, au moment où les emplois respectables étaient rares pour les femmes.

AN EXHIBITION of Samuel Butler's photography in Gallery Sixteen, an elegant rotunda room just off the entrance to the Tate Britain, offered a rare opportunity to see some of the photography of the author of Erewhon and to contemplate how Victorian photographic realism fares in the setting of a modern museum. The exhibition, celebrating the centenary of Butler's death, ran from November 2002 to May 2003 and was made up of thirty-five framed photographs, some of them digitally touched up by Dudley Simons, and an assortment of photobooks and editions of Butler's self-illustrated volumes. It was developed by Tate curator Richard Humphreys and Butler scholar Elinor Shaffer, with the support of librarian Mark Nicholls from St. John's College at Cambridge, which houses most of Butler's extensive photographic work in its special collections. Titled “Samuel Butler and the Ignorant Eye,” after Shaffer's notion in her Erewhons of the Eye: Samuel Butler as Painter, Photographer, and Art Critic (1988) that Butler's photography renders “the eye of the viewer … ignorant and open” (229), the black-and-white secularism of Butler's work offered a startling change in imagery from the intense colorism of “Rossetti and Medievalism,” the exhibit that preceded it in Gallery sixteen.

Elio Modigliani collected images from the island of Nías, which he then published in A journey to Nías (1890). Many of these images refer to materials held and exhibited by the Museum of Anthropology and Ethnology of the University of Florence. Here we conducted some trials on the use of artificial intelligence (AI) not only to colorize them, but also to retrieve information from the images useful for restoration and conservation. Modigliani's photographs are of particular interest. They illustrate and integrate his reflections on the use of photography for anthropological field research, offer valuable information on shooting techniques and on the original condition and colors of the artifacts. Although AI may not always reconstruct with absolute fidelity the original colors, especially for clothing, our results provides important guidelines for analysis and allows a cogent discussion of the possibilities of AI technology in anthropological research.

Colorism is a historical and cultural phenomenon with global dimensions. This phenomenon is credited, in part, for the growth and financial success of the skin-lightening industry throughout the world. In Thailand, dark skin is often associated with lower-income classes, including farm labor. As a pioneering effort, this study sought to empirically identify colorism in a way that would minimize social desirability bias. Three groups in an MBA program at an international university in Bangkok - Thais with Chinese ancestry, non-Chinese Thais, and non-Thai foreigners – were asked to identify socio-economic status (“rich” or “poor”) by examining photographs of male and female Thais with dark, medium, and light skin tones. An analysis was conducted using the demographic variables of ethnicity, gender, and age. The study found significant differences wherein Thai-Chinese associated light skin tone with wealth and dark skin tone with poverty. Non-Chinese Thais followed in this assessment, but foreigners did not. Regarding marriage, both Thai-Chinese and non-Chinese Thais chose photographs of individuals with light skin to marry, but they were significantly different than the foreigners where skin tone was not a factor. There was no significant difference regarding gender. However, age was significant in that the younger group in the study (ages 20 to 27) identified dark skin tones as indicating poverty more than the older group in the study (ages 28 to 48).

The public expected color from the early photographic images, yet, daguerreotypes—the first commercially available photographic process—failed to register the natural colors. Daguerreotypists developed several coloring methods to solve this inconvenience after 1840. Scientific analyses of the hand-colored daguerreotypes are limited, and the primary information sources available are manuals and patents. This study aims to contribute to the knowledge of hand-coloring techniques, which impacts conservation practices, mainly the cleaning procedures and complements technical art history investigations. We studied nine colored daguerreotypes with a non-invasive methodology based on three spectroscopic techniques: X-ray fluorescence (XRF), micro-Raman spectroscopy (μ-Raman), and micro-Fourier-transform infrared spectroscopy in reflection mode (μ-rFTIR). The results revealed the different colorants and some of their mixtures employed by the colorist. It also adds information regarding the photographic production of three studios: Désiré François Millet, active in Paris between 1840 and 1868; Antoine Claudet (1797–1867), active in London; and the James E. McClees and Washington Lafayette Germon studio, operative between 1846 and 1855 in Philadelphia, USA. These technical details constitute a helpful comparison to future studies on Daguerreian studios and colored daguerreotypes.

Abstract: Modification of art may be viewed as enhancement or vandalization. Even though for a long time many were opposed to the idea of colorizing images, they now have finally viewed it for what it is - an enhancement of the art form. Grayscale image colorization has since been a long-standing artistic division. It has been used to revive or modify images taken prior to the invention of colour photography. This paper explores one method to reinvigorate grayscale images by colorizing them. We propose the use of deep learning, specifically the use of convolution neural networks. The obtained results show the ability of our model to realistically colorize grayscale images. Keywords: Deep Learning, Convolutional Neural Network, Image Colorization, Autoencoders.

Following World War II, Ebony's creator and editor, John H. Johnson, sought to create a popular black magazine in the vein of Life and Look that would reflect the accomplishments and joys, “the happier side,” of African American life.1 Throughout the first four years of its publication, Lena Horne appeared on the magazine's cover three times – the only woman to do so during this period. In this paper, I argue that the fledgling Ebony magazine drew on Lena Horne's wartime status as a beautiful black icon and represented her as a symbol of its ideological project, broadly, and as the Ebony image of postwar black womanhood, specifically. The magazine's representation of Lena Horne acts as a useful trope for understanding how Ebony imaged postwar black femininity in terms of motherhood, work, and civil rights activism; additionally, Ebony's representation of Horne and Ebony readers' letters to the editor reveal central issues of respectability, pinup photography, colorism, hair care, and interracial relationships as they were debated within the magazine's pages.Behind the lavish make-up, gay tinsel and brilliant glitter of American's most popular Negro entertainer, Lena Horne is a wonderfully human, somewhat lonesome, amazingly-honest, militant-minded personality who is relatively unknown to a vast audience of millions of movie, radio, and night club fans.2

Color affects the psychological and physiological state of a person and is one of the powerful factors that form a comfortable visual environment. Currently, color-leaved taxa are actively offered by nurseries and are used in landscaping. At the same time, the color characteristics of taxa are very diverse, are quite subjective, which makes their selection and application difficult. The purpose of the work is to analyze and classify the modern varietal variety of the wood assortment for the middle part of Russia by the color of ripened leaf blades. The analysis was carried out according to descriptions and photographs presented in information sources and catalogues of nurseries. It has been established that the modern coloristic diversity of arboreal taxa is great, there are discrepancies in the description of foliage color, there is no classification. The classification of the colorist of taxa by visual perception, convenient from a practical point of view, has been developed, reduced to the main color groups: green, blue (silver), whitish, whitish-pink, yellow, purple. The peculiarity is that variegated taxa are included in these groups in accordance with their color perception: yellow-mottled and yellow–bordered taxa are classified as yellow, taxa with white inclusions and fringing are whitish, and so on. It can be recommended to nursery farms to apply the developed classification when selling planting material, landscape firms and gardeners when selecting and composing plant compositions.

Abstract A remarkable quality of Chinese vernacular photographs during the socialist period is their efflorescence of color. From studio colorists well-versed in the art of potassium cyanide to private brushstrokes in thick marker, the prevalence of photo-coloring (zhuose) evidenced a marked degree of dissatisfaction with an unmanipulated print. Bracketed by Chinese photo studios’ national institutionalization in 1956 and the promulgation of commercial color rolls in the 1980s, this discussion builds from extant scholarship on Chinese vernacular photographs of the 1960s and 1970s. By analyzing colored Tiananmen photographs of the Cultural Revolution, a posthumous biography of photo-colorist Zhou Qi, and various examples from the volume Anonymous Photos (Yiming zhao), this article demonstrates that photo-coloring was a crucial aesthetic agent in negotiations between officially mandated postures and bold ways of inscribing individual experiences as chromatic repositories of memory and meaning, marking a site where mass visuality and more specific imaginations of the self collided into color.

A long legacy of media imagery persistently distorts, stereotypes, and ignores marginalized racial and ethnic groups despite widespread calls to diversify media representations. In particular, fashion and beauty media continue to feature light-skinned models and celebrities over dark-skinned individuals, even lightening dark skin with photo editing to achieve ideals of whiteness and lightness. This practice aligns with colorism, or the privileging of light skin tones for access to economic and social capital. This study examines colorism in a particular genre of digital photography, online retail images, as a problem of visual representation. The novel method of visual computational analysis is used to quantitatively compare how mainstream clothing retail brands represent model skin tones across still and video media modes. The findings suggest that analyzed retailers tended to favor light-skinned models on their websites and that model skin tones in product videos were significantly darker than in product photos. These findings are considered through research on race and technology, photographic manipulation, and media misinformation. Ultimately, the study suggests that visual (in)consistencies can reveal the role of structural biases in shaping media representations. The article also provides a methodological tool for conducting this work.

La colorisation est le processus qui consiste à ajouter des couleurs aux images en niveaux de gris. C’est une tâche importante dans la communauté de l’édition d’images et de l’animation. Bien que des méthodes de colorisation automatique existent, elles produisent souvent des résultats insatisfaisants en raison de défauts tels que le débordement de couleur, l’incohérence, des couleurs non naturelles et la nature non trivial du problème. Par conséquent, une intervention manuelle est souvent nécessaire pour obtenir le résultat souhaité. En conséquence, il y a un intérêt croissant à automatiser le processus de colorisation tout en permettant aux artistes d’ajouter leur propre style et vision. Dans cette thèse, nous étudions divers formats d’interaction en guidant les couleurs sur des zones spécifiques d’une image, ou en les transférant à partir d’une image ou d’un objet de référence. Nous introduisons deux méthodes de colorisation semi-automatiques. Tout d’abord, nous décrivons une architecture d’apprentissage profond pour la colorisation d’images qui prend en compte les images de référence de l’utilisateur. Notre deuxième méthode utilise un modèle de diffusion pour coloriser des dessins en utilisant des indications de couleur fournies par l’utilisateur. Cette thèse commence par l’état de l’art des méthodes de colorisation d’images, des espaces de couleur, des métriques d’évaluation et des fonctions de perte. Bien que les méthodes de colorisation récentes basées sur des techniques d’apprentissage profond obtiennent les meilleurs résultats, ces méthodes sont basées sur des architectures complexes et un grand nombre de fonctions de perte, ce qui rend difficile leur compréhension. Pour cela, nous utilisons une architecture simple afin d’analyser l’impact de différents espaces de couleur et fonctions de perte. Ensuite, nous proposons une nouvelle couche d’attention appelée super-attention qui utilise des superpixels. Elle permet d’établir des correspondances entre les caractéristiques hautes résolutions de paires d’images cible et référence. Cette proposition permet d’atténuer le problème de la complexité quadratique des couches d’attention. De plus, elle aide à surmonter les défauts de débordement de couleur dans la tâche de colorisation. Nous étudions son utilisation pour le transfert de couleur, et pour la colorisation basée sur des exemples. Nous proposons également une extension de ce modèle afin de guider spécifiquement la colorisation sur des objets segmentés. Enfin, nous proposons un modèle de diffusion probabiliste basé sur des conditionnements implicites et explicites, pour apprendre à coloriser des dessins au trait. Notre approche permet d’ajouter des interactions utilisateur à travers des indices de couleur explicites tout en s’appuyant sur l’entraînement du modèle de diffusion principal. Nous utilisons un encodeur spécifique qui apprend à extraire des informations sur les indices de couleur fournis par l’utilisateur. Ce modèle permet d’obtenir des images colorisées diverses et de haute qualité
Colorization is the process of adding colors to grayscale images. It is an important task in the image-editing and animation community. Although automatic colorization methods exist, they often produce unsatisfying results due to artifacts such as color bleeding, inconsistency, unnatural colors, and the ill-posed nature of the problem. Manual intervention is often necessary to achieve the desired outcome. Consequently, there is a growing interest in automating the colorization process while allowing artists to transfer their own style and vision to the process. In this thesis, we investigate various interaction formats by guiding colors of specific areas of an image or transferring them from a reference image or object. As part of this research, we introduce two semi-automatic colorization frameworks. First, we describe a deep learning architecture for exemplar-based image colorization that takes into account user’s reference images. Our second framework uses a diffusion model to colorize line art using user-provided color scribbles. This thesis first delves into a comprehensive overview of state-of-the-art image colorization methods, color spaces, evaluation metrics, and losses. While recent colorization methods based on deep-learning techniques are achieving the best results on this task, these methods are based on complex architectures and a high number of joint losses, which makes the reasoning behind each of these methods difficult. Here, we leverage a simple architecture in order to analyze the impact of different color spaces and several losses. Then, we propose a novel attention layer based on superpixel features to establish robust correspondences between high-resolution deep features from target and reference image pairs, called super-attention. This proposal deals with the quadratic complexity problem of the non-local calculation in the attention layer. Additionally, it helps to overcome color bleeding artifacts. We study its use in color transfer and exemplar-based colorization. We finally extend this model to specifically guide the colorization on segmented objects. Finally, we propose a diffusion probabilistic model based on implicit and explicit conditioning mechanism, to learn colorizing line art. Our approach enables the incorporation of user guidance through explicit color hints while leveraging on the prior knowledge from the trained diffusion model. We condition with an application-specific encoder that learns to extract meaningful information on user-provided scribbles. The method generates diverse and high-quality colorized images