Video Portraits Research Articles

Towards imbalanced motion: part-decoupling network for video portrait segmentation

Towards imbalanced motion: part-decoupling network for video portrait segmentation

Coverage of fighting tasks of ukrainian aviation in documentaries

Introduction. Documentaries about Ukraine's aerospace industry create conditions for the audience to immerse themselves in the world of aviation, to feel the effect of being present during combat missions, to cover military events and military operations, but in the first place, they tell the story of a person or people. Documentary filmmakers touched upon the topic of aviation at the beginning of the Russian-Ukrainian war in 2014 and 2015. After the full-scale invasion, the role of aviation became more important in the war, and a new wave of documentaries about military aviation appeared. The peculiarities of covering the combat missions of Ukrainian aviation during the full-scale invasion of Ukraine by the Russian Federation, the use of the latest methods of recording events, transmitting testimonies, and creating images ofheroes in documentaries are the subject of this study. The purpose of the study is to identify the features of contemporary television documentaries on aviation in the context of the Russian-Ukrainian war. Research methods. The results of the study were obtained using theoretical, descriptive, comparative methods, and systematic monitoring of documentaries about Ukrainian military aviation. The features of these films were identified, in particular, by the following criteria: duration, methods of recording events and facts, peculiarities of presentation, creation of characters, artistic techniques. The documentaries from 2015 and 2023 were compared. Results. We have analysed documentaries about the combat missions of Ukrainian aviation, such as Pilots. Return", "Masters of the Sky", "Downed Russian pilots". The authors of the films use the techniques of video portraits, video interviews, and comments by experts or heads of relevant structures. Testimonies and interpretations of events help the viewer to understand the specifics of aviation. Conclusions. Documentary films enhance the publicistic images of heroes, including pilots and other aviation professionals, Ukrainian defenders, not only by recording facts, documenting events, testimonies, comments and interviews, but also by artistic techniques and editing.

Real-time high-resolution video portrait matting network combined with background image

Real-time high-resolution video portrait matting network combined with background image

Memories are One-to-Many Mapping Alleviators in Talking Face Generation.

Talking face generation aims at generating photorealistic video portraits of a target person driven by input audio. According to the nature of audio to lip motions mapping, the same speech content may have different appearances even for the same person at different occasions. Such one-to-many mapping problem brings ambiguity during training and thus causes inferior visual results. Although this one-to-many mapping could be alleviated in part by a two-stage framework (i.e., an audioto- expression model followed by a neural-rendering model), it is still insufficient since the prediction is produced without enough information (e.g., emotions, wrinkles, etc.). In this paper, we propose MemFace to complement the missing information with an implicit memory and an explicit memory that follow the sense of the two stages respectively. More specifically, the implicit memory is employed in the audio-to-expression model to capture high-level semantics in the audio-expression shared space, while the explicit memory is employed in the neural-rendering model to help synthesize pixel-level details. Our experimental results show that our proposed MemFace surpasses all the state-of-theart results across multiple scenarios consistently and significantly.

Ageing and Care in Agnès Varda

Abstract This article analyses selected works of Agnès Varda as original contributions to cultural gerontology. It focuses on Varda’s pronounced interest in older people throughout her career and explores how, in experimental portraits and self-portraits, she approaches ageing by harnessing surprising devices — temporal, spatial, aesthetic, haptic, and affective — to challenge ageism and assumptions of disability, and to encourage identification. The article pays particular attention to a series of interlinked video portraits of a fourth-age female nude in 7 p., cuis., s. de b… (à saisir) (1984) and to Varda’s bold and innovative exploration and performance of her own old age in her co-directed film Visages villages (2017). The latter foregrounds consistently the physical and affective dimensions of old age and may be read not only as an extended self-portrait, but more broadly as a protracted exercise in empathy with older people. The article examines how the film cuts across thematic and aesthetic norms both by foregrounding older subjects and by validating them as fascinating and beautiful. It also reads Visages villages as a powerful appeal to an intergenerational ethics of care, established in part through Varda’s creative partnership with street photographer and artist JR.

Robust Video Portrait Reenactment via Personalized Representation Quantization

While progress has been made in the field of portrait reenactment, the problem of how to produce high-fidelity and robust videos remains. Recent studies normally find it challenging to handle rarely seen target poses due to the limitation of source data. This paper proposes the Video Portrait via Non-local Quantization Modeling (VPNQ) framework, which produces pose- and disturbance-robust reenactable video portraits. Our key insight is to learn position-invariant quantized local patch representations and build a mapping between simple driving signals and local textures with non-local spatial-temporal modeling. Specifically, instead of learning a universal quantized codebook, we identify that a personalized one can be trained to preserve desired position-invariant local details better. Then, a simple representation of projected landmarks can be used as sufficient driving signals to avoid 3D rendering. Following, we employ a carefully designed Spatio-Temporal Transformer to predict reasonable and temporally consistent quantized tokens from the driving signal. The predicted codes can be decoded back to robust and high-quality videos. Comprehensive experiments have been conducted to validate the effectiveness of our approach.

UniFaceGAN: A Unified Framework for Temporally Consistent Facial Video Editing.

Recent research has witnessed advances in facial image editing tasks including face swapping and face reenactment. However, these methods are confined to dealing with one specific task at a time. In addition, for video facial editing, previous methods either simply apply transformations frame by frame or utilize multiple frames in a concatenated or iterative fashion, which leads to noticeable visual flickers. In this paper, we propose a unified temporally consistent facial video editing framework termed UniFaceGAN. Based on a 3D reconstruction model and a simple yet efficient dynamic training sample selection mechanism, our framework is designed to handle face swapping and face reenactment simultaneously. To enforce the temporal consistency, a novel 3D temporal loss constraint is introduced based on the barycentric coordinate interpolation. Besides, we propose a region-aware conditional normalization layer to replace the traditional AdaIN or SPADE to synthesize more context-harmonious results. Compared with the state-of-the-art facial image editing methods, our framework generates video portraits that are more photo-realistic and temporally smooth.

Photorealistic Audio-driven Video Portraits.

Video portraits are common in a variety of applications, such as videoconferencing, news broadcasting, and virtual education and training. We present a novel method to synthesize photorealistic video portraits for an input portrait video, automatically driven by a person's voice. The main challenge in this task is the hallucination of plausible, photorealistic facial expressions from input speech audio. To address this challenge, we employ a parametric 3D face model represented by geometry, facial expression, illumination, etc., and learn a mapping from audio features to model parameters. The input source audio is first represented as a high-dimensional feature, which is used to predict facial expression parameters of the 3D face model. We then replace the expression parameters computed from the original target video with the predicted one, and rerender the reenacted face. Finally, we generate a photorealistic video portrait from the reenacted synthetic face sequence via a neural face renderer. One appealing feature of our approach is the generalization capability for various input speech audio, including synthetic speech audio from text-to-speech software. Extensive experimental results show that our approach outperforms previous general-purpose audio-driven video portrait methods. This includes a user study demonstrating that our results are rated as more realistic than previous methods.

Instinct Extinct: The Great Pacific Flyway

This article, or artist’s inquiry, was written in concert with the exhibition Instinct Extinct: The Great Pacific Flyway. Beginning with introductions to bird migration, the concept of global flyways and the history of conservation, the text considers the poetics of art-making relative to academic research. Areas of artistic exploration include a map depicting California’s changing waterscape, video portraits of people of the flyway and assemblages of invented and found avian artifacts. The article concludes with a review of current environmental conditions affecting migratory birds and some reflective passages.

American Dreams ft. David Lynch

American Dreams ft. David Lynch Niels Niessen (bio) We’re like the dreamer who dreams and then lives inside the dream. But who is the dreamer? —Twin Peaks: The Return (created by Mark Frost and David Lynch, 2017, U.S.) “Hello, you’re tuned into Interview Project. Today we’re meeting Suzy. Suzy is Daniela’s aunt. Enjoy the interview.” Or: “Hello, you’re tuned into Interview Project. Today we’re meeting Darryl. Darryl is waiting to buy a ranch in Montana. Enjoy the interview.” We’re tuned into Interview Project, a web-based series of 121 video portraits directed by Austin Lynch and Jason S. The videos are about four minutes each and they were posted online every three days between June 2009 and Click for larger view View full resolution Figure 1. David Lynch presents: [End Page 31] Click for larger view View full resolution Figure 2. The image scrambles while tuning in on Lynch. Click for larger view View full resolution Figure 3. Welcome to Interview Project. May 2010. Together they document the road trip Lynch, Jason S., and their team made across a small-town United States that remains unseen by the interstate traveler. It is a portrait of a nation reminiscent of John Steinbeck (The Grapes of Wrath, Travels with Charley) as well as Walter Evans’s American Photographs of the Great Depression, and it is in the vein of those Great-American works that Interview Project captures the dilapidated state of the American dream. [End Page 32] TUNING INTO AMERICA Interview Project was produced by Absurda, the production company of David Lynch, Austin’s father, and the project is also hosted on David Lynch’s website: interviewproject.davidlynch.com. Although Lynch—David, that is—did not go along on the trip, he does introduce each of the videos, in his own Lynchian way, and above all in the same way each time. We press “play” and the scrambling image makes it seem like we’ve just powered on one of those old-school color televisions from the time color television was still a new medium. Once the image has stabilized on a close-up of his face, Lynch welcomes the viewer: “Hello, you’re tuned into Interview Project. Today we’re meeting Spira” (Figures 1–3). As always, Lynch’s spirits are high. He is seated in front of a concrete wall and a yellow chest of drawers, while he’s wearing a blue suit and a white, tieless shirt that is completely buttoned up. His silver hair is combed yet messy and contrasts nicely with the wall. Cut to a medium-long, awkwardly high-angled shot that shows him sitting at a giant desk in a space that Lynch followers will recognize as his Los Angeles studio, from where for a while he also broadcast his odd daily weather reports, and which also forms the setting for the 2016 documentary David Lynch: The Art Life (d. Jon Nguyen, Rick Barnes, U.S./Denmark). Like so many others in this online era, Lynch has been working a lot from home lately. On his desk—messy-arranged as Lynch himself—there is an ashtray, a box of drawing utensils, a red shiny ball the size of an apple, a steaming cup of coffee (brewed, we assume, from his own signature blend), while a ray of sunlight casts all of this in a homely glow. “Spira used to dance to The Doors at the Whisky-a-Go-Go.” Cut to the original camera position. Lynch smiles: “Enjoy the interview,” upon which the intro dissolves into a disclaimer that “the views expressed in these interviews are those of the participants alone and do not necessarily reflect the views of Interview Project or its sponsors.” The portraits themselves are utterly sincere slices of American life. Take episode 3, which shows Kee, whom the team met in a Navajo reservation in Tuba City, Arizona (Figure 4). “Hello, my name is K, K.J., I go by K.J., but my name is Kee Jackson.” Kee is framed in close-up against the backdrop of the red rocks of the Arizona desert. His mouth twitches, he seems a little nervous...

The Unlovable Press

The saving grace of journalism. This is what Michael Schudson called objectivity in a video portrait shown during the ceremony in which he was awarded an honorary degree at the University of Gronin...

Martyrs for the Mass

“Martyrs for the Mass” takes the position of the vicarious witness (witness to the image of witnessing), and attempts to contemplate how we (along with the thousands of others who visited St. Paul’s Cathedral on 23 June 2016) confront, encounter, and, in our case, ingest video portraits.

The VideoMob Interactive Art Installation Connecting Strangers through Inclusive Digital Crowds

VideoMob is an interactive video platform and an artwork that enables strangers visiting different installation locations to interact across time and space through a computer interface that detects their presence, video-records their actions while automatically removing the video background through computer vision, and co-situates visitors as part of the same digital environment. Through the combination of individual user videos to form a digital crowd, strangers are connected through the graphic display. Our work is inspired by the way distant people can interact with each other through technology and influenced by artists working in the realm of interactive art. We deployed VideoMob in a variety of settings, locations, and contexts to observe hundreds of visitors’ reactions. By analyzing behavioral data collected through depth cameras from our 1,068 recordings across eight venues, we studied how participants behave when given the opportunity to record their own video portrait into the artwork. We report the specific activity performed in front of the camera and the influences that existing crowds impose on new participants. Our analysis informs the integration of a series of possible novel interaction paradigms based on real-time analysis of the visitors’ behavior through specific computer vision and machine learning techniques that have the potential to increase the engagement of the artwork's visitors and to impact user experience.

I Me Mine: Artistic Self/Artistic Persona

This work comprises of a series of video portraits and sound recordings that explore the connections between gesture, gaze and voice in video portraiture. Most of the subjects are artists represented in the current exhibition, but the disjunction between sound and image makes it difficult to categorically identify the authors of the sound bites on the work’s soundtrack. Designed to function as an introduction to the themes and issues generated by the Self/Persona relationship in the nascent field of Persona Studies, the work is also concerned with the presentation of the artistic self, and the ‘loss of self’ that may or may not occur as a consequence of artistic practice. Formally, the piece plays with the repetition and symmetry to underscore the vulnerability and mutability of the self within contemporary culture.

Aesthetics of self-scaling: parallaxed transregionalism and Kutluğ Ataman's art practice

This article examines relations of ethnography, contemporary art practice, globalisation and scalar geopolitics with particular reference to Kutluğ Ataman's artworks. Having been shortlisted for the Turner Prize at the Tate and awarded the prestigious international Carnegie Prize in 2004 with his 40-screen video installation Küba (2004), Ataman became an extremely well-known, globally acclaimed artist and filmmaker. Self-conscious of their global travel and critically attentive to the contemporary ethnographic turn in the visual arts scene, Ataman's video works perform a conscientious failure of representing cultural alterity as indigeneity. Concentrating on the artist's engagement with ethnography, this article consists of three main parts. Analyses of the selection of videos in each part will give an account of different scalar aspects of Ataman's artworks. It will first revisit a previous study (Çakirlar 2011) on the artist's earlier work of video portraits including Never My Soul! (2002) and Women Who Wear Wigs (1999). A detailed discussion of Küba follows, a work that exemplifies a scalar transition in his critical focus – from body and identity to community and geopolitics. The discussion then moves to a brief analysis of the series Mesopotamian Dramaturgies, including the screen-based sculptures Dome (2009), Column (2009), Frame (2009), English as a Second Language (2009) and The Complete Works of William Shakespeare (2009). Rather than addressing scale as a differential concept, this article aims to demonstrate the ways in which Ataman's art practice produces self-scaling, self-regioning subjects that unsettle the hierarchical constructions of scale and facilitate a critique of the scalar normativity within the global art world's regionalisms and internationalisms.

‘I am an American’: protesting advertised ‘Americanness’

How are citizenships and nationalisms constructed, connected, and contested in the post-9/11 USA – performatively, affectively, and visually – and how do their relationships figure ‘Americanness’? This article takes up this question (1) by tracking how Americanness was advertised in the American Ad Council's ‘I am an American’ campaign and (2) by introducing the multimedia project ‘I am an American’: Video Portraits of Unsafe US Citizens, which engages the Ad Council's campaign as a practice-based protest of the Ad Council's advertised ‘Americanness’. The article traces how the Ad Council's campaign advertises what Evelyn Alsultany calls ‘diversity patriotism’. It also constructs a complex, mobile system of differentiation that marks some citizens as ‘safe’ and others as ‘unsafe’, which runs counter to the idealized notion of a unified ‘Americanness’ that it advertises. The article then examines how the practice-based protest project ‘I am an American’ takes these ‘unsafe citizens’ – US citizens who either will not or cannot make their differences normatively conform to the national ideal of the ‘One’ composed of the ‘Many’ propagated by the Ad Council's campaign – as its point of departure to reflect upon how citizenship protests function for and against citizenship, nationalisms, and various figurations of Americanness.

Queue Portrait: Ang Cui

Ang Cui is a Ph.D. student at Columbia University in New York City. His research focuses on embedded devices such as routers, printers and VOIP phones. He is the inventor of a novel, host-based defense mechanism known as Symbiotes. Symbiotes are designed specifically to retrofit black-box, vulnerable, legacy embedded systems with sophisticated anti-exploitation mechanisms. In this video portrait, Ang describes the extent of the embedded threat in real-world environments, discusses novel exploitation techniques for embedded systems--like enterprise networking equipment--and develops practical defenses for embedded systems that constitute our global communication substrate.

The photo in the Ffilm: public and private collections in video-portrait

This article discusses the inclusion of photographs in ethnographic films, particularly in the genre video portrait. In the reconstitution of an individual's history, photographic images play an important role in the evocation of past facts that often remain only as fragments of memory. When examining personal collections and public archives, we prospect for photographic and iconographic images that allow usto construct possible relationships between collective and individual memories.

Science Lives: Video Portraits of Great Mathematicians

Science Lives: Video Portraits of Great Mathematicians

The 14th Congress of the International Society for Twin Studies: Selected Highlights/Research Summaries: Twin Study of Telomere Erosion; Genetics of Monozygotic Twinning; Developmental Dysplasia; Fertility of Mothers of Twins / General Interest: Video Portraits; Separated at Birth; Childhood Actors; Adult Twin Film Villains

Selected highlights from the 14th International Congress on Twin Studies, set in Florence, Italy in April 2012, are presented. This meeting, which traditionally occurred every three years, is now being held every two years. Reviews of research on the topics of stress and aging (telomere erosion), the genetics of MZ twinning, developmental dysplasia, and fertility of mothers of twins follow. The final section includes several public interest items, namely an exhibit of twins' video portraits, unusual physical look-alikes, and young and adult twin television and film actors.