Recently, pure transformer-based models have shown great potentials for vision tasks such as image classification and detection. However, the design of transformer networks is challenging. It has been observed that the depth, embedding dimension, and number of heads can largely affect the performance of vision transformers. Previous models configure these dimensions based upon manual crafting. In this work, we propose a new one-shot architecture search framework, namely AutoFormer, dedicated to vision transformer search. AutoFormer entangles the weights of different blocks in the same layers during supernet training. Benefiting from the strategy, the trained supernet allows thousands of subnets to be very well-trained. Specifically, the performance of these subnets with weights inherited from the supernet is comparable to those retrained from scratch. Besides, the searched models, which we refer to AutoFormers, surpass the recent state-of-the-arts such as ViT and DeiT. In particular, AutoFormer-tiny/small/base achieve 74.7%/81.7%/82.4% top-1 accuracy on ImageNet with 5.7M/22.9M/53.7M parameters, respectively. Lastly, we verify the transferability of AutoFormer by providing the performance on downstream benchmarks and distillation experiments. Code and models are available at https://github.com/microsoft/Cream.
As an important step towards visual reasoning, visual grounding (e.g., phrase localization, referring expression comprehension/segmentation) has been widely explored Previous approaches to referring expression comprehension (REC) or segmentation (RES) either suffer from limited performance, due to a two-stage setup, or require the designing of complex task-specific one-stage architectures. In this paper, we propose a simple one-stage multi-task framework for visual grounding tasks. Specifically, we leverage a transformer architecture, where two modalities are fused in a visual-lingual encoder. In the decoder, the model learns to generate contextualized lingual queries which are then decoded and used to directly regress the bounding box and produce a segmentation mask for the corresponding referred regions. With this simple but highly contextualized model, we outperform state-of-the-arts methods by a large margin on both REC and RES tasks. We also show that a simple pre-training schedule (on an external dataset) further improves the performance. Extensive experiments and ablations illustrate that our model benefits greatly from contextualized information and multi-task training.
María Guadalupe Valladares González, Giselda Emérita Hernández
El trabajo aborda una visión crítica de las autoras sobre el viaje cognitivo realizado por la maestranda Libety Martínez en la Universidad de las Artes de Cuba. Desde una perspectiva de distanciamiento se analiza el camino identitario recorrido por ésta, (Martínez/2024) para corregir limitaciones y normalizaciones impuestas por la comunidad científica en la antedicha universidad basadas en epistemocentrismos y etnocentrismos disciplinares. Este trabajo va a los estratos más remotos de estas construcciones investigativas que responden en alguna medida a relaciones de poder.
En el contexto de la educación artística contemporánea, la práctica artística no solo constituye un medio de expresión, sino también una vía de investigación y una forma de conocimiento. La figura animal, como componente esencial del lenguaje visual, ha ocupado históricamente un lugar central en la creación artística. Desde su papel como tótem que simboliza la naturaleza y lo divino, ha evolucionado hasta convertirse en un medio para reflejar estructuras sociales, experiencias individuales y la conciencia educativa. En la pintura moderna y contemporánea, los significados culturales y las estrategias visuales asociadas a las imágenes animales han superado ampliamente la mera representación, convirtiéndose en un lenguaje narrativo visual con potencial crítico y constructivo.Este estudio se basa en la práctica pictórica con el objetivo de explorar cómo la imagen animal puede convertirse en un canal para la imaginación educativa y la reflexión social. A través del análisis del lenguaje visual y los sistemas simbólicos presentes en las obras de artistas como Goya, Antonio Saura, Picasso, Francis Bacon y Anselm Kiefer, se identifican características formales relacionadas con las manchas y los residuos visuales, los cuales se toman como punto de partida para la creación personal. Sobre esta base, se construye un sistema educativo ficticio centrado en figuras animales específicas —el Bestiario Educativo— que combina creación artística, construcción de personajes y crítica pedagógica. Esta propuesta intenta responder, mediante recursos visuales, a los problemas estructurales de la educación contemporánea y proponer una posibilidad educativa diversa y orientada al futuro.En esta investigación, la pintura no solo actúa como medio de expresión visual, sino también como herramienta de indagación práctica. A diferencia de las metodologías dominadas por la fotografía en la investigación educativa artística, la pintura ofrece una dimensión más procesual, emocional y generativa. A través de la transformación visual de la mancha —desde lo figurativo hacia lo abstracto— se ha desarrollado un método creativo que incluye la construcción de personajes educativos, el desarrollo de sistemas narrativos y la proyección de imaginarios futuros, convirtiendo a la pintura en una construcción educativa en sí misma, más allá del contenido didáctico.
The emergence of artificial emotional intelligence technology is revolutionizing the fields of computers and robotics, allowing for a new level of communication and understanding of human behavior that was once thought impossible. While recent advancements in deep learning have transformed the field of computer vision, automated understanding of evoked or expressed emotions in visual media remains in its infancy. This foundering stems from the absence of a universally accepted definition of “emotion,” coupled with the inherently subjective nature of emotions and their intricate nuances. In this article, we provide a comprehensive, multidisciplinary overview of the field of emotion analysis in visual media, drawing on insights from psychology, engineering, and the arts. We begin by exploring the psychological foundations of emotion and the computational principles that underpin the understanding of emotions from images and videos. We then review the latest research and systems within the field, accentuating the most promising approaches. We also discuss the current technological challenges and limitations of emotion analysis, underscoring the necessity for continued investigation and innovation. We contend that this represents a “Holy Grail” research problem in computing and delineate pivotal directions for future inquiry. Finally, we examine the ethical ramifications of emotion-understanding technologies and contemplate their potential societal impacts. Overall, this article endeavors to equip readers with a deeper understanding of the domain of emotion analysis in visual media and to inspire further research and development in this captivating and rapidly evolving field.
Musawenkosi M. Tapfuma, Regis Musavengane, Rowen Magwaza
ABSTRACT With emerging forms of tourism, creative tourism has brought a shift in the tourism industry. Creative tourism is not fully embraced and there is paucity of studies on the new type of tourism in the developing world. The focus of this article is arts and crafts as a form of creative tourism and its role in promoting inclusive tourism from a community-based type of inclusive tourism. Using qualitative approaches in Zimbabwe cities, the study examines the role of creative tourism through visual arts and crafts in promoting inclusive tourism. The key findings of the study indicated that when managed well, visual arts and crafts have great potential in promoting inclusive tourism through the reduction of poverty and improvement of the livelihoods of the marginalised communities. However, artists encounter challenges in the production of arts and crafts that include lack of capital, lack of production space, lack of marketing, rivalry products from other countries, and lack of support from the government and local tourism organisations. Despite the challenges faced, artists appeal for strategies in financing and marketing, fair competition and relaxed government policies.
ABSTRACTPurpose We implemented and evaluated a hybrid 4-week arts-based elective for clinical medical students to support flourishing.Materials and Methods Five students participated in early 2022. Twelve sessions occurred in-person at art museums and other cultural centers, and five occurred online. Sessions incorporated varied arts-based learning activities, including Visual Thinking Strategies, a jazz seminar, and a mask-making workshop. We evaluated the course via weekly reflective essays, interviews 6 weeks after the course, and pre-post surveys that included four scales with clinical relevance: capacity for wonder (CfW), tolerance for ambiguity (TFA), interpersonal reactivity index, and openness to diversity.Results Qualitatively, the course helped learners: 1) reconnect with individual characteristics and interests that had been neglected during medical education; 2) better appreciate others’ perspectives; 3) develop identities as physicians; and 4) engage in quiet reflection, renewing their sense of purpose. Quantitatively, pre-post mean totals increased for the CfW (32.0 [SD 6.8] vs 44.0 [SD 5.7], p=.006) and TFA scales (16.4 [SD 5.2] vs 24.2 [SD 6.9], p=.033).Conclusions This elective facilitated learners’ connecting with themselves, others, and their profession with improvement in clinically-relevant measures. This provides further evidence that arts-based education can foster professional identity formation and be transformative for students.
Cinematographers are trained to control and measure the relative difference in brightness between two parts of a scene, or a face, for expressive purposes. Painting is often referred to for inspiration as practitioners learn to compose and represent light and shadow in an aesthetically considered manner. In this respect, it is noteworthy that the painters generally studied by film-making students are mostly from Renaissance traditions and produced work featuring predominantly White models. This gap of racial representation in cinematographic pedagogy is stark and has long been overlooked. This article mounts an enquiry into the lack of diversity in cinematography education, examining how different aesthetic traditions, such as Asian ink paintings, could pave new ways for decolonising the conventional conceptions of lighting ratios. Drawing on qualitative and quantitative case studies undertaken in university workshops, I discuss how students respond to Black and Asian artwork as visual references when they are tasked with lighting models with non-White skin tones to accommodate the different reflectance of their skins. By comparing the learning outcomes and current industry techniques for optimising screen representation of Black, Asian and minority ethnic groups, the article evaluates how students respond to learning from modern artwork that promotes diverse identities, and argues for the benefits of integrating greater inclusiveness into cinematography.
Representing altered states of consciousness, even through the most phantasmal of technical images, is an inherent contradiction; once we attribute a physical body, i.e. objectivity, to mental images, we deny what Husserl considers their very essence. Jean Mitry draws from this assumption when discussing filmic access to mental states from a phenomenological perspective. The following essay reconsiders Mitry’s contribution with specific reference to the role of projection, technically and metaphorically speaking, in the cinematic technique and imagination; this, with the intention of suggesting some crucial questions for the comparison between the filmic forms of the visible and those inaugurated by the technology of the virtual.
Nineteenth-century English stained glass has produced a rich corpus of works mainly produced by the many large and well-known Victorian studios. One group of artisans who can add much to the current debate on nineteenth-century stained glass are amateurs. However, almost every scholarly publication discussing stained glass or nineteenth-century material culture has ignored this minority group. Driven by the rejection of mass-produced windows that predates the well-known innovations of the Arts and Crafts Movement, this article is part of ongoing research and it discusses the cultural and social significance of nineteenth-century amateur windows in Anglican churches. Influenced by the Gothic Revival and the ideologies of the Oxford Movement and the Ecclesiologists, the range of amateur individuals is extremely diverse and most had a close connection with the Anglican Church. Inspired by religious literature, women amateurs, in particular, played an important sociocultural role striving for the demands of paid employment, training, education, and access to the professions. A small selection of case studies will explore some of the discovered amateur windows. The article will raise arguments in the context of Victorian religious material and visual cultures and will discuss the art of personalized making.
Chek Lap Kok (Hong Kong Airport) 21.00 01.12.19, a short video by Stephen Connolly which documents a walk to Hong Kong Airport from the Expo centre on the airport island, by means of slow travel, under makeshift conditions, and without carbon expenditure. The video offers a brief exploration of the materialities and grounded infrastructures of aviation at a moment of pandemic-led change and invites us to look anew at the familiar and banal physical geography of the airport and how we move within it, drawing on Lefebvre’s Production of Space and theories of ‘Spatial Cinema’.
3D model retrieval is becoming a hot research topic due to its wide applications such as computer-aided design, digital entertainment, and virtual reality. For this challenging task, feature learning and similarity measure are two critical problems. However, existing approaches usually learn discriminative visual features and develop a complex graph matching strategy to measure the similarity independently. In this paper, we propose an unsupervised method which can embed similarity measure into the feature space. The proposed method utilizes both similarity and dissimilarity information to better leverage the unsupervised problem and estimates the labels which are further used for metric learning. With the learned metric, we project the original features to more discriminative feature space and efficiently measure the similarity among models under the new feature space. We conduct extensive evaluations of three popular and challenging datasets. The experimental results demonstrate the superiority and effectiveness of the proposed method, competing against the state of the arts.
Esse texto aborda o tema da aprendizagem e do ensino de desenho a partir de uma vivência pessoal. Tem o objetivo de compartilhar experiências sobre um contexto específico, passível de comparações com outros casos desse campo de estudos. A partir da perspectiva autoetnográfica de David M. Hayano (1979), constrói uma narrativa biográfica cronológica. Apresenta uma síntese descritiva de eventos cotidianos relacionados à problemática do desenvolvimento de habilidades técnicas do desenho figurativo. Descreve algumas das dificuldades encontradas durante as atividades de ensino de desenho em cursos de graduação na área de artes visuais. Relaciona algumas das estratégias utilizadas para superá-las, e finaliza com uma reflexão sobre o papel das repetições e correções na aquisição destas competências.
En la revista de arte Artforum de junio de 1966 Robert Smithson (New Jersey, 1938) escribe un inquietante y radical ensayo que llevaba por título Entropía y los nuevos monumentos. El ensayo analiza el trabajo de una serie de artistas y las concepciones implicadas en la producción de los creadores citados en el mismo; también de cómo se inscriben en un determinado contexto espaciotemporal tales producciones. Smithson desarrolla una serie de conceptos, que han sido tratados tanto por él como por otros autores y artistas, de una manera intensa e interesante en términos conceptuales, poéticos y plásticos, pero a la vez con un rigor científico calibrado de una manera exótica, extraña y difusa, pero muy sugerente e inspiradora. Entre los conceptos empleados se encuentra el de entropía, que es a la vez sencillo y enigmático, pero también complejo y esencial. El concepto de entropía es relevante en términos semióticos para comprender las interacciones inevitables entre diferentes campos de conocimiento y la estructuración tanto del lenguaje como de la acción comunicativa y su deriva. En el presente artículo se analizan algunos de los conceptos y definiciones empleados por Smithson y otros autores, artistas o no, y la relevancia que estos tienen para el trabajo con imágenes, la producción de sentido a través de estas, y el constante retroajuste (retrolectura, etc…) que debemos aplicar para evitar, o al menos ralentizar, la desaparición (grado de visibilidad) o fuga de las imágenes constituidas o conformadas, independientemente de su naturaleza. Es decir, se intenta investigar la tendencia, a través de la generación y traslación de matemas físico matemáticos, entre otras herramientas, de las pretendidas imágenes a la fuga tanto de su potencial sentido lingüístico como plástica, visualmente (su desaparición, alejamiento o momento de equilibrio), y los factores que hacen de las imágenes y su producción, emisión y recepción, objetos de conocimiento independientes y liberados con un alto potencial metafórico, y por consiguiente, enlazados con conceptos y dinámicas tales como entropía, incertidumbre y gravitación.
Extendemos un sincero saludo a la comunidad académica del diseño, su constante apoyo ha permitido avanzar en el proceso de fortalecimiento de nuestra publicación. Cada día aumenta el número de colaboradores que desde diferentes países se integran a este proyecto de divulgación académica de las investigaciones que sobre el diseño, la comunicación, la imagen y áreas afines se adelantan en instituciones universitarias de la geografía iberoamericana. El fascículo 18 de la Revista KEPES. Grupo de Estudio en Diseño Visual reúne parte de estos resultados, 12 artículos que presentamos a nuestros respetados lectores.
Salnikova Ekaterina V. / Сальникова Екатерина Викторовна
The article is devoted to the socio-cultural phenomenon of the television Studio and its theatrical and literary background. Partly this research can be attributed to the archeology of media, more precisely, to the archeology of modern television formats aesthetics. One of the most typical television programmes is considered - the combination of Studio conversation and broadcast screen works of different genres. The author supposes that the most important elements of Studio programs are their regularity and controllability, as well as the atmosphere of safe distance from the big world or from many virtual and real worlds. Civilized, planned communication and management of information streams is modeled in the Studio. Therefore, violations of this pattern make such a strong impression.
The origins of the necessity for a Studio atmosphere are largely associated with the ancient culture and, to an even greater extent, Renaissance images and moods – the existence in the “charming corner” (locus amoenus), the search for the lost Paradise harmony, awareness of the high value of the spectator and commentator security.
…
In the Internet era videoblogger’s home becomes the Studio and synthesizes privacy and virtual publicity. The idea of a secure public space of communication no longer prevails, giving way to the idea of total integration of the local "place of action" into the space of infinite real and virtual worlds, which should increase the socio-cultural status of the private territory and its inhabitants.
Статья посвящена социокультурному феномену телевизионной студии и его театрально-литературной предыстории. Частично данное исследование может быть отнесено к археологии медиа, точнее, к археологии эстетики современных телевизионных форматов. Рассматривается один из наиболее типичных видов телевизионного вещания — совмещение студийного действа и трансляции экранных произведений различных жанров в рамках единой программы. Автор полагает, что важнейшими элементами студийных программ являются их упорядоченность и контролируемость ведущими, а также атмосфера безопасной дистанцированности от большого мира или от множества виртуальных и реальных миров. В студии моделируется цивилизованное общение и управление информационно-образными потоками. Коммуникация рационализирована, спланирована. Поэтому столь сильное впечатление производят нарушения этого паттерна.
Истоки потребности в студийной атмосфере во многом связаны с античными и, в еще большей степени, ренессансными образами и настроениями — пребыванием в «прелестном уголке» (locus amoenus), поисками утраченной райской гармонии, а также развитием индивидуализма и осознанием высокой ценности защищенности человека-зрителя и человека-комментатора. Неслучайно именно в процессе развития ренессансных форм рождается театр, в котором зритель дистанцирован от разыгрываемого представления и не может быть подвергнут никакому воздействию со стороны исполнителей. Неслучайно в трех наиболее показательных для Ренессанса сборниках новелл Чосера, Боккаччо и Маргариты Наваррской созданы своеобразные обрамления для отдельных рассказов – смоделированы ситуации многодневного рассказывания историй и их комментирования в приятном обществе. Телевизионные студийные программы унаследуют функцию безопасного для участников и зрителей, занимательного и эстетически привлекательного действа – наблюдения за большим миром и размышления о нем вслух в светском обществе собеседников. В ХХ в. комфортный, лишенный прозаических бытовых черт интерьер, разновидностью которого является телестудия, занимает место центрального образа идеального публичного пространства.
В эпоху интернета комната-студия видеоблогера синтезирует в себе приватность и виртуальную публичность. Идея безопасного публичного пространства общения уже не главенствует, уступая место идее тотальной интегрированности локального «места действия» в пространство бесконечных реальных и виртуальных миров, что должно повысить социокультурный статус приватной территории и ее обитателя.
Computational censorship in the form of fake news and toxic comments regulation is a subject that comes up quite often in the public discourse, as a result of the volatile political circumstances on a global scale and due to the unquestionable impact of journalism on these circumstances. Public attention has been directed to the role of mainstream and other media in the formation of public opinion, either in the form of articles or in the form of usergenerated comments. The purpose is to analyse and allow a deeper understanding of a project that is under development, namely, computational-censorship and to show that algorithmic regulation is not a solution, but rather another layer to a more fundamental problem.
This article examines the implicationsof developing Machine Leraning/Artificial Iintelligence (ML/AI) which aims to regulate the internet and we attempt to allow a glimpse into the technical aspect of the problem as a way to back arguments that could be re-jected by the ML/AI research community as “non-pragmatic”. Finally, it aims to highlight the absurdity of the current approach to research in this area, which is the exact opposite of the rationalism that the field claims to be embracing.