Week 1 

What is a Deepfake?

• Key words: deep learning, fake, manipulated, machine learning, artificial intelligence, algorithms 

A combination of “deep learning” and “fake”, deepfakes
are hyper-realistic videos digitally manipulated to
depict people saying and doing things that never
actually happened (CNN03; FRB04).

Introduction

Deepfakes are digitally manipulated media, replacing one person’s likeness with another. They use machine learning and artificial intelligence techniques to create visual and audio content that can be easily deceived. Image forensics develops techniques to detect manipulated images.

 

How does it work? How is it believable? 

Deepfake is an AI technology that uses machine learning algorithms to create convincing hoax images, sounds, and videos, replacing real person’s images or voices with artificial likenesses.

List some examples with images.

 

Use Of Deepfake In Different Industries: pornography, media, films, social media, Tv

List keywords associated with the topic :-

• Deepfake.
• Artificial Intelligent.
• Algorithms.
• Digitally manipulated.
• machine learning methods
• Ethics
• Copyright
• DE aging
• Motion Capture
• Law
• Different Platforms

Is it okay to replace actors’ faces using deep fake technology?

I believe that with the consent of the actors/actresses or taking their rights to use in films and doing so in ethical ways for entertainment purposes utilizing deepfake technology is ok, but it is not ok if it is used without the actor’s full approval there is a risk of potential copyright infringement and ethical concerns when deepfakes are used. For example we have seen deep fake technology been used for movies like, Digital Version of Paul Walker in Fast & Furious 7, Younger Robert De Niro in “The Irishman” and Younger Double of Sean Young in “Blade Runner 2049”. Deepfake technology has the potential to transform the film business in a variety of ways. Movies might be made without the requirement for stars to be physically present on set by using digital reproductions of real actors, lowering the vast costs associated with filmmaking and also  by allowing filmmakers to construct totally new stories or bring historical figures back to life. Furthermore, 

Deepfake technology, when used ethically and with consent from actors, can transform the film industry by reducing costs, allowing filmmakers to create new stories, and bringing historical figures back to life. However, using it without full consent raises copyright infringement and ethical concerns. Examples include Fast & Furious 7, “The Irishman,” and “Blade Runner 2049.”

 

Search for sources on Deepfakes?

In March 2021, TikTok users fooled deepfake detection software by presenting realistic deepfake videos of Tom Cruise, using over 13,000 images from various angles, demonstrating the complexity of such videos(Jeong et al., 2021).

Havard References

Refrenses:

Jeong, S., Sturtevant, M. and Stephen, K. (2021) Responding to deepfakes and disinformation , The Regulatory Review. Available at: https://www.theregreview.org/2021/08/14/saturday-seminar-responding-deepfakes-disinformation

 

Week 2: AI developments and VFX. The VICO research framework

A.I use in VFX.

Question: How is AI changing visual effects production, and what are the challenges and potential advantages of AI in this context?

 

• Is AI a challenge or advantage?
• Types of AI?
• What specific examples can you find of AI use with visual effects?

• Briefly write a sentence or two on how you think think AI might change visual effects?

Ai is transforming the world of visual effects. AI in visual effects is a artistic tool which helps or enhance the creative process and  its still in his initial stage so till near future its a tool to support the creative process. It has its own advantages and limitation. AI can automate VFX production but may struggle to understand the intricate creative decisions made by artists, requiring artistic creativity and intuition. Filmmakers and directors rely on VFX artists to bring their projects to life, but AI may struggle to interpret and adapt to these individualized artistic intents. It also have issues with Ethical and Cultural Considerations and Technical Challenges.

Type of Ai like Chat GPT, Midjourney, Descript, Steve AI: Script to Videos, Runway, Movavi Video Editor for Quick Edits

Deep Fakes are AI programs that use Convolutional Neural Networks (CNN), auto encoders, and Generative Adversarial Networks (GAN) to create fake videos. GAN is used by Facebook for image, video, and voice generation. CNN is used in automated self-driving cars and AI vision in robotics. Auto-encoders simplify and copy information from pre-existing code, removing unnecessary data, and can also be used to create adult videos.

In 2015, Belgium-based ScriptBook developed software to predict movie success, plan, and recommend films based on script analysis. The software also considers likability and potential profit potential. (Using AI in the VFX and Film Industry | Kate Xagoraris | Ontario)

KOGNAT is used for rotoscopey  (R0tomation)

SwitchLight (beeble.ai)

 

 

What’s in the Proposal

500-600 words:

• Prepare and informally present your initial ideas for your proposal :- Title “Exploring the Efficacy/Result of AI and Conventional Techniques in Matte-Painting”

 

• Example topic / research activity :-   AI Software Like: Midjourney, Runway, Pikalabs.

• Review some past assignments:-

 

• Work on the proposal:-

 

• Discuss what you aim to find out :-
• The area of research with a question
• Aims / objectives / keywords :-  AI, VFX COMPOSITING, MATTEPAINTING, DIGITAL PAINTING, COMPOSITE, PROMPTS, ALGORITHMS. MACHINE LEARNING, Mathematical Equations,

 

 

“Comparative Analysis of AI and Traditional Methods in Matte-Painting”

Questions: What are the major breakthrough in Matte-Paintings for films? / What is the role of AI in creating matte- painting? / Does AI is Efficient in Creating Matte- painting? / Is AI just an another tool in this Journey? /Journey of Matte-painting with innovation of new tools and techniques/

Aims/objectives/keywords: – AI, VFX COMPOSITING, MATTE PAINTING, DIGITAL PAINTING, COMPOSITE, PROMPTS, ALGORITHMS. MACHINE LEARNING, Mathematical Equations,

 

1. An Introduction to the Investigative Study

I will investigate the AI-generated matte painting in visual effects comparing it with traditional and digital methods and highlighting its benefits and potential drawbacks. Matte-painting – Firstly, I will write about what is Matte-painting? Use Matte-painting and traditional methods such as hand-painting and digital manipulation techniques. Secondly, Explain artificial intelligence (AI) and its expanding impact on VFX industries. Examine how machine learning and AI algorithms/Prompts are used to create a matte painting. Talk about the benefits of AI-powered matte painting, including its increased speed, efficiency, and realism. Examine how AI algorithms can analyze and understand visual elements, textures, and lighting to create realistic matte paintings. Highlight the ability of AI to learn from existing matte paintings and generate new compositions based on specific requirements. what are the Challenges. Collaboration between AI and Artists.

I am going to look at really important text by Manovich (2023) that emphasizes the historical importance of various creative tools, such as Adobe Photoshop and 3D modelling software, in encouraging artistic expression and media production. It emphasises the ongoing evolution of AI-driven creative tools, demonstrating how features like Photoshop’s ‘Magic Wand’ and ‘Quick Selection’ tools, which were once thought to be AI-driven, have now become standard functionalities people now event don’t notice or considered as AI. The discussion situates these advances within a broader trajectory of simulating human artistic abilities, demonstrating the constant attempt to push creative boundaries and reshape cultural perceptions of AI.  (Manovich, 2023).

2. Including the aims and objectives

The Aim and objective of this investigation is to understand the difference and relation  between AI and Traditional matte painting process. Objective is to find out is AI with Human Creativity or AI against humans creativity or both but it requires human touch.

3. A Methodology or list of five key sources (References) – these should be annotated.

My primary research method will be practical, in which I will create (seek out live brief from the client) live brief in response to which I will create two matte paintings. one will be using existing techniques such as digital brushes(toolset) I will source diff material and images for the first version. secondly  I will create matte- matte painting using generative fill or prompt AI. tell AI to create matte-painting according to the brief. Then consider which results are more likely to meet the brief. I will analyze the process and steps. I will also be looking for the secondary resources such as book and  I will investigate matte-painter who are working with Ai. Second, I can look for pros and cons, and techniques used in creating matte painting composites, and ask questions such as, “Is this technique or AI efficient for creating or in the process of creating the entire film?” What method can I use in my case? I’ll do it practically. Is AI an assistant or just a tool, or does it feel alien? How good is AI? Can it understand feelings or emotions? Can it replicate emotion/tone/feel or is it very plain? I will give my opinion on the result and keep it open for public. is it visually descriptive.

 

Manovich, L. (2023) Lev Manovich, Lev Manovich – Artificial Aesthetics: A Critical Guide to AI, Media and Design. Available at: http://manovich.net/index.php/projects/artificial-aesthetics (Accessed: 22 October 2023).

4. Any important images as figures

Here are the example of matte-painting creating using AI generative fill in photoshop.

In support of the references

Literature Review and Analysis 

 

Artificial Aesthetics: A Critical Guide to AI, Media and Design  Chapter 5 – AI images and Generative Media. 

Manovich, L. (2023) ‘Chapter 5 – AI images and Generative Media.’, in Artificial Aesthetics: A Critical Guide to AI, Media and Design. Available at: http://manovich.net/content/04-projects/168-artificial-aesthetics/lev-manovich-ai-aesthetics-chapter-5.pdf (Accessed: 30 October 2023).

(Manovich, 2023)

It’s says that  it was deeply historicist: rather than inventing everything from scratch, it innovated by adapting certain older aesthetics to contemporary art contexts. 

The majority of the millions of regular people and creative professionals that use generative media tools only use them in their current form. This could alter in the future if networks built using our own data become more user-friendly. However, notwithstanding these details, every recently manufactured cultural object made by There is a common logic for trained nets. Generic media artifacts are not like conventional paintings, drawings, or sculptures. made from the ground up. Furthermore, they are not the outcome of recording any kind of sensory phenomena, like images, movies, or audio files. Rather, they are constructed from AI visuals and digital media – 8 A vast collection of additional media artifacts. This creative process connects creative media to older artistic methods and genres. 

  Going further back in time, we find a broad cultural paradigm that was also a reaction to the accumulation of historical art and culture artifacts in easily accessible media collections. This is paradigm is known as “post-modernism.”  What about “modernism” in the 1910s and 1920s? While the overall emphasis was on originality and novelty, one of the procedures it developed in search of novelty was direct quotations from the vast universe of contemporary visual media that was rapidly expanding at the time in response to this visual intensification of mass culture, in the early 1910s Georges Braque and Pablo Picasso began incorporating actual newspaper, poster, wallpaper, and fabric fragments into their paintings. A few years later, John Heartfield, George Grosz, Hannah Hoch, Aleksandr Rodchenko, and a handful of other artists began to develop photo-collage techniques. Photo-collage became another method of creating new media artifacts from existing mass media images. 

I am going to look at really important text by Manovich (2023) that emphasizes the historical importance of various creative tools, such as Adobe Photoshop and 3D modelling software, in encouraging artistic expression and media production. It emphasizes the ongoing evolution of AI-driven creative tools, demonstrating how features like Photoshop’s ‘Magic Wand’ and ‘Quick Selection’ tools, which were once thought to be AI-driven, have now become standard functionalities people now event don’t notice or considered as AI. The discussion situates these advances within a broader trajectory of simulating human artistic abilities, demonstrating the constant attempt to push creative boundaries and reshape cultural perceptions of AI.  (Manovich, 2023). 

 

AI for Arts  Niklas Hageback, Daniel Hedblom

With the advent of industrialism in the early twentieth century, machines began to replace our biological capabilities with the manual labor they could marshal; today, artificial intelligence is beginning, and has already made significant progress, in replacing the second essential component of the human composition, namely our mind.

The invention of the camera is an example of how photography evolved into its own art form over time. But it also shaped paintings, maybe pulling them away from prior genres that emphasized exactness, as a camera could now portray this much better. Or where we are now in the technological progression where digital editing, such as Photoshop, allows us to express, or mislead, subjects with qualities that were before impossible. Furthermore, technological advancements have permitted

Method and Methodology

Method and methodology, I will be doing practical work as well as secondary research on AI generative AI and its impact on matte painter or painted backdrops. In recent years, HD led video walls have been introduced, which provide realistic backdrops with the Unreal engine. However, with generative AI, a new option for generating images or backgrounds has emerged, such as in Photoshop and many other AI such as Cuebric, which uses prompts to generate images based on stable diffusion model creation in painting, editing, up scaling, and exporting. Historically, filmmakers used painting on glass, physical sets, miniatures, and digital sets, but with the introduction of AI generated environments, this has changed.So I’ll be creating AI-generated matte-paintings in Photoshop using prompts or tools. and the other I will create using the traditional method, compare it, describe the process, and open to the public for feedback and criticism. So we can consider whether it is merely a tool, a threat to jobs, or a new revolution.

 

“Comparative Analysis of AI and Traditional Methods in Matte-Painting”

 

INTRODUCTION

Matte painting has played an important role in bringing imaginary worlds to life in the ever-changing landscape of filmmaking, where imagination meets technology. Matte painting has come a long way, from precise brushstrokes on glass to the arrival of computer sets. However, the most recent protagonist in this story is not a talented painter holding a brush, but the computational capability of Artificial Intelligence. Consider a film set where artificial intelligence creates magnificent landscapes, surpassing the limitations of physical objects or computer effects. This enthralling promise is not limited to science fiction; it is a transformational reality that is transforming the future of cinema (pholmesphd, 2021)

It will be comparative research, which pits AI-generated matte paintings against its conventional counterparts at the confluence of tradition and innovation. As we investigate the impact of AI on this cinematic art form, we wonder whether it is simply a tool, a threat to jobs/Employment, or the beginning of a cinematic revolution. Manovich (2023) that emphasizes the historical importance of various creative tools, such as Adobe Photoshop which were once thought to be AI-driven, have now become standard functionalities people now event don’t notice or considered as AI. The discussion situates these advances within a broader trajectory of simulating human artistic abilities, demonstrating the constant attempt to push creative boundaries and reshape cultural perceptions of AI.  (Manovich, 2023).

The Aim and objective of this investigation is to understand the difference and relation between AI and Traditional matte painting process. Objective is to find out is AI with Human Creativity or AI against humans’ creativity or both, but it requires human touch.

In this essay, we’ll explore the history of matte painting, going from traditional methods with glass to today’s AI-generated matte paintings. We’ll look at the pros of both traditional and digital matte paintings, highlighting the special touch and creativity artists bring. Moving on to Artificial Intelligence (AI), we’ll talk about where it came from, how it evolved, and the computer-based instructions (algorithms) that make it work. We’ll dig into the idea of AI prompts, questioning if AI is just following instructions or if it truly has creativity. We’ll weigh the good and not-so-good sides of AI, thinking about whether it’s more of a helper for artists or if it might replace jobs by doing tasks automatically. The essay will also check if having a human touch is crucial in the creative process, asking about the role of humans working with AI. An important point we’ll explore is giving credit for art – should it go to the artist, the AI, or both? This leads to a bigger discussion about how AI and human creativity come together, making us think about the possible teamwork or conflicts between them in the creative field. To see how generative AI works in matte painting, the essay will end with making a matte painting using Photoshop. This practical part aims to show how AI can create what we want based on a given plan, giving us a clearer picture of how artists and AI can work together in making visual art.

Traditional matte painting is like an artist’s fingerprint, unique and touchable. But here comes AI, offering speedy tricks and fresh ideas. We’ll see what each brings to the creative table, like a friendly competition between old-school and new school.  So, get ready to step into the world of matte painting and AI wonders. It’s like exploring two sides of creativity – one that started with a paintbrush, glass, digital tools like photoshop and the other with algorithms and brainy tech. Let’s see how these two worlds dance together in our Essay.

 

Literature Review

Generic media artifacts are not like traditional paintings, drawings, or sculptures, but rather constructed from AI visuals and digital media. They connect creative media to older artistic methods and genres. Post-modernism, a cultural paradigm that emerged in response to the accumulation of historical art and culture artifacts, emphasizes originality and novelty. In the 1910s and 1920s, modernism focused on originality and novelty, incorporating contemporary visual media into paintings. Georges Braque and Pablo Picasso began incorporating newspaper, poster, wallpaper, and fabric fragments into their paintings in the early 1910s. John Heartfield, George Grosz, Hannah Hoch, Aleksandr Rodchenko, and others developed photo-collage techniques, creating new media artifacts from existing mass media images.

Manovich (2023) emphasizes the historical importance of creative tools like Adobe Photoshop and 3D modeling software in encouraging artistic expression and media production. The ongoing evolution of AI-driven creative tools, such as Photoshop’s ‘Magic Wand’ and ‘Quick Selection’ tools, has led to the development of new functionalities that are now considered standard functionalities. These advances are part of a broader trajectory of simulating human artistic abilities, pushing creative boundaries and reshaping cultural perceptions of AI. (Manovich, 2023).

AI is increasingly being used in various fields, including text to video, where tools like Meta’s “Make-a-Video” and Google’s Imagen AI can create silent films from sentences. Hugging Face, an AI organization, developed its own open-access text to video tool, which created a viral film using prompts like “Will Smith eating spaghetti.” The film was created using a Modelscope Text2Video generator, which converts text prompts into small video segments. The film was made worse when the word “meatballs” was added to the prompt, resulting in Smith’s smiling head being placed on a horrifying “body” made of a throbbing pile of meatballs. Big directors like Baz Luhrmann have used AI to blend Elvis face with Austin, believing that AI assistants will be like mobile phones. However, AI struggles to remember the average human’s fingers and can make weird hands. For example, Henry Cavalry cannot shave as he grows for another role, requiring CGI that costs around 25 million dollars. AI can also be used in sitcoms like Deep Fake Neighbor Wars, where actors can be given faces. Keanu Reeves, who has been fighting Agents the Matrix for decades, requested a clause in contracts stating that his performance won’t be digitally manipulated. (Saunders,2023). With the advent of industrialism in the early twentieth century, machines began to replace our biological capabilities with the manual labor they could marshal; today, artificial intelligence is beginning, and has already made significant progress, in replacing the second essential component of the human composition, namely our mind.

The invention of the camera is an example of how photography evolved into its own art form over time. But it also shaped paintings, maybe pulling them away from prior genres that emphasized exactness, as a camera could now portray this much better. Or where we are now in the technological progression where digital editing, such as Photoshop, allows us to express, or mislead, subjects with qualities that were before impossible. Furthermore, technological advancements have permitted (Hageback and Hedblom, 2021).

Joanna Zelenska’s text ai art: machine visions and warped dreams criticises the question “Can computers be creative?” as misguided, highlighting the superficiality of AI-driven art, which frequently prioritises aesthetics over meaningful creativity. It criticises the emphasis on imitating existing styles, especially in projects such as “style transfer,” which imitate established artistic canons. The discussion calls into question traditional notions of authorship, originality, expertise, and taste, urging a reconsideration of human creativity as partially computational. It proposes viewing art as a collaborative outcome of human and nonhuman agents within various technical systems, rather than as a strict division between humans and machines. The text advocates for a post-humanist approach to art history, redefining the human role within the larger apparatus of creation.

Artificial Intelligence and the Arts: Toward Computational Creativity Article from the book The Next Step: Exponential Life by (Bolter, 2016) Computational creativity involves creating software that exhibits human-like creative behaviour, such as inventing theories, writing poetry, painting, and composing music. This field not only creates autonomous creative systems, but it also aids in the understanding of human creativity and the development of collaborative tools. Despite skepticism in the past, computational creativity has matured, as evidenced by increased activity, sophisticated software, and cultural value. Debates include whether the Turing test is appropriate for evaluating creative software and the difficulty of admitting that computers are not human. The article contends that, while creativity is often regarded as mysterious, it can be studied and replicated using AI techniques. Creativity is defined as a novel and valuable combination of known ideas, with musical and visual arts examples provided.. The paper concludes with reflections on the democratization of creativity through human-assisted AI systems.

The Article AI tools can create new images, but who is the real artist? O’Brien, M. and Lajka, A. (2023) states that AI tools, including DALL-E, Midjourney, and Stable Diffusion, have reached a point where they can rapidly generate new images, mimicking the styles of renowned artists like Van Gogh. However, this advancement has prompted legal action from living artists and photographers who claim that AI software companies are reproducing and manipulating millions of copyrighted images without proper licencing. Getty Images, the developer of Stable Diffusion, has initiated legal proceedings against Stability AI, accusing the London-based startup of infringing intellectual property rights for commercial gain. Another lawsuit, filed in a US federal court, claims that AI-generated images compete in the market with original works. This legal backlash highlights the challenges and ethical concerns raised by AI tools capable of producing visual media, as well as concerns about misinformation and public trust. In the context of matte painting, these developments highlight the growing intersection between AI and artistic creation, prompting discussions on copyright, artistic attribution, and the role of AI in the creative process.

The book the artist in the machine: the world of AI-powered creativity by Miller, Arthur I. Investigates how AI-powered computers are becoming more creative in art, literature, and music, potentially surpassing human achievements. Arthur I. Miller, a creativity expert, investigates the factors that influence the creative process and interviews experts in artificial intelligence. He exhibits computer-generated art, such as terrifying creatures and music composed by a machine. While acknowledging computers’ existing creative abilities, Miller recognises the need for AI to enter the world in order to properly replicate human creativity. Despite addressing a future in which computers may outperform humans in terms of originality, the book ultimately embraces AI’s positive creative potential in a variety of artistic fields.

Method and Methodology:

My primary research method will be practical, in which I will create (seek out live brief from the client). The brief is to create the Christmas advert John Lewis the snowman’s journey in response to which I will create two matte paintings. To accomplish this, I will employ Adobe Photoshop as my primary tool for creating matte paintings, ensuring a comprehensive exploration of both conventional and cutting-edge methods. The initial step in the traditional approach involves curating diverse materials and images to construct the matte painting. This process will focus on the mastery of digital brushes and a specific toolset within Photoshop. The emphasis will be on the artistic interpretation and technical skills of the matte painter, relying on established methods to bring the vision to life. This hands-on, artist-driven exploration of the creative process aims to showcase the capabilities of traditional techniques within the Photoshop environment. secondly, I will create matte- matte painting using generative fill or prompt AI. tell AI to create matte-painting according to the brief. Then consider which results are more likely to meet the brief. By instructing the AI system based on the brief, I will explore the capacity of AI to understand and execute complex creative tasks within the Photoshop software. This method raises questions about the adaptability and potential efficiency of AI in meeting the demands of a specific creative project, all within the familiar Photoshop interface. I will analyze the process and steps. I will also be looking for the secondary resources such as book and I will investigate matte painter who are working with Ai. The research extends beyond the technical aspects to explore broader questions related to AI in creative processes. Delving into the pros and cons of traditional and AI-driven techniques, I aim to understand their respective efficiencies and limitations. Crucial inquiries will be posed, such as the suitability of AI for crafting entire films and whether it functions as a mere assistant or a distinctive creative entity. The exploration will also touch upon the experiential aspect – does working with AI feel collaborative or does it introduce an element of foreignness into the creative process? To provide a comprehensive overview, I will assess the emotional intelligence of AI. Can it understand and replicate human emotions, tones, and feelings, or does it produce outcomes that lack depth and emotional resonance? My research will culminate in an open-ended discussion of the results, presenting my personal opinions while inviting public engagement to enrich the discourse on the visual descriptive quality of AI-generated matte paintings.

 

 

Findings and Discussion

 

In this chapter, we compare the results of AI-created matte paintings with those created using traditional filmmaking methods. We investigate how realistic, creative, and efficient these approaches are, as well as the complex relationship between AI technology and human artistic abilities. We discuss what these findings mean for the future of matte painting in the film industry by combining information from relevant studies and practical research.

Matte painting has been used in film since 1907, when Norman Dawn created the first recorded example for “California Missions.” Originally, scenes were painted on glass and placed in front of the camera (Mattingly, 2011, p.16. Rickitt, 2010, p.244). However, due to the weight of the glass and the need for quick painting to minimise exposure, this posed challenges. Dawn pioneered the ‘original negative matte painting’ method in 1911, in which live-action areas were exposed on location, leaving space for later matte paintings. The use of blacked-out areas on glass and the attachment of black cards to the camera lens were both innovations. Despite technological advances such as the Optical Printer and digital tools in the late 1980s, Dawn’s’matting’ principle persisted.

Figure 1

 

Norman Dawn’s (Figure 1) self-portrait sketch shows his tented workstation with Debrie camera, and the glass shot used to rebuild the historic crumbling ‘Mission La Soledad’ in Monterey, California.

With the rice of Digital age, A matte painter named Garrett Fry noticed that directors now prefer more exciting scenes with cool camera angles. With the rise of digital technologies in filmmaking, People are wondering if the old way of matte painting is still good with new computer stuff. But even though we have fancy 3D computer tools, matte painting is still liked. the issue therefore becomes, why is matte painting still as relevant and popular in the production of digital environments for cinema and television as it has always been? Dylan Cole, a famous contemporary matte painter with IMDB credits on titles such as Avatar (2009), Maleficent (2014), Tron Legacy (2010), Golden Compass (2007), and the Lord of the Rings Trilogy (2001 – 2003), responds to some extent to this. In a 2012 interview with 2D Artist Magazine, he identifies three main types of matte painting 2D, 2.5D (camera projection), and full 3D matte painting. The key benefits of matte painting are control and ease, particularly for one-off shots or scenes with limited perspectives and camera movements (pholmesphd, 2021). A matte painting is a flat image of a scene that works well for static shots but struggles with moving cameras. The 2.5D approach attempted to address this by employing techniques such as zooming, but it was unable to provide true depth information. Disney’s Multiplane Camera assisted with parallax shifts, but it had limitations. Digital tools improved this by allowing for faster and more flexible matte painting application onto geometric surfaces. However, problems persisted until the late 1990s, when VFX studios began to use 3D software like Maya for more accurate representations, seamlessly aligning CG sets with live footage. Consider Photoshop in this day and age. The idea is that such these tools are just there to enhance what you already bring to the table in terms of creativity; for example, Photoshop still requires a significant amount of effort to create anything unique.

Matte paintings do not need to be realistic; they simply need to complement the overall look of the film. The purpose of matte painting is to supplement the film’s look or concept development. Matte painting, at its core, is about recreating an atmosphere that was either not present during production, was too expensive to capture in real life, or was simply impossible to shoot for various reasons. A matte painter, also known as a matte artist, creates the illusion of something that appears real but does not exist in reality by painting a – typically static – background or other film world components. This could be the interior of a massive star ship in one scenario, or the hazy backdrop of H. G. Wells’ industrial London in another (Using impossible software to create impossible worlds – the rise of AI and the art of digital painting. | Dazzle Pictures, 2023).

What exactly is AI (Artificial Intelligence)? In our context, AI refers to computer systems that analyse massive amounts of data, apply algorithms, and generate new creations based on this data. This includes digital art for VFX studios and other creative endeavours. The emphasis here is on AI-generated art, in which images are created solely using AI applications, a topic that has recently received a lot of attention. As with any new technology, opinions on AI in art differ. On the one hand, some see it as a threat to artists’ livelihoods, echoing concerns raised during the introduction of digital cameras in photography. Critics argue that the minimal effort required on the input side contrasts sharply with the sophisticated output, raising questions of fairness—suggesting that art is being created without the traditional labour-intensive process (Using impossible software to create impossible worlds – the rise of AI and the art of digital painting. | Dazzle Pictures, no date).

The arrival of AI heralds a new era in which the digital sphere merges with artistic expression. AI-driven visual effects, powered by machine learning and neural networks, have ushered in a wave of innovation that pushes the bounds of animation well beyond what was previously thought possible (HOUND STUDIO, 2023).

Generative Artificial Intelligence, also known as Generative AI, is a game-changing technology that uses artificial intelligence to create novel and unprecedented objects or effects. This type of AI uses advanced algorithms to learn from data, allowing it to produce novel and distinct results. Unlike traditional programming, where specific instructions are given for output, generative AI can generate conclusions and outputs on its own using its reasoning and decision-making abilities. The efficiency of generative AI in generating new content outperforms that of manual coding, making it a highly valuable tool for businesses looking to improve customer engagement. This technology’s potential applications are virtually limitless, opening up new avenues for innovation and creativity. (The dawn of generative ai: a threat to creatives or a boon? 2021).  Image Generation or Generative AI models have the capability to craft entirely new images with just a few input parameters. This technology finds applications in digital art and photo manipulation, where it can generate realistic or surreal images without the need for extensive time and effort. Thanks to open-source tools like DALL-E, Stable Diffusion, Midjourney, and Lexica, businesses can now easily harness the power of this technology for various creative purposes (The dawn of generative ai: a threat to creatives or a boon? 2021).

Neural networks in generative AI can create new texts and visuals at a level comparable to skilled writers, artists, or photographers after being trained on massive amounts of text and images from the internet. This technology, which is similar to the human brain in complexity, operates on trillions of connections between artificial neurons rather than fixed algorithms. Current generative AI systems, such as GPT and Stable Diffusion, are trained on large and diverse datasets, but they can also be tailored to specific cultural areas or artists from different historical periods. The AI art project of the Refik Anadol Studio, for example, used neural networks trained on tens of thousands of artworks from the MoMA collection, representing a century of modern art (1870 – 1970). Despite the apparent disparity between modernism and AI training processes, modern artists frequently drew inspiration from past art traditions, much like neural networks learn from historical culture and art. In essence, while modern artists rejected certain artistic paradigms of the time, they did not reject art history entirely. They innovated by adapting old aesthetics to new contexts, similar to how neural networks learn from the past to create something new. The connection between modern art and experimental psychology was also important, as artists in the 1910s used images from psychology studies to explore human visual sensation and perception. The process of creating with generative AI involves building on a vast archive of existing media artifacts, drawing parallels with historical art creation methods like film editing or composite photography (Manovich, 2023)..

Movements in art history such as post-modernism and modernism drew inspiration from accumulating cultural artefacts, forming new artistic expressions from existing references. Contemporary artificial intelligence artworks, such as Unsupervised or Pereulkov’s Artificial Experiments 1-10, carry on this tradition by employing neural networks trained on cultural databases. By navigating through patterns and regions in the universe of contemporary art, these works explore and expand the possibilities of art and its techniques. (Manovich, 2023).

 

AI, or artificial intelligence, is a controversial tool, with some seeing it as a threat to creativity. However, we contend that AI is simply another tool, similar to Photoshop, designed to enhance rather than replace existing talent. AI, particularly in the form of matte painting, has become an essential part of the visual arts workflow. Matte painting is the process of creating backgrounds or environments that would be too expensive or impossible to film in real life. While traditional matte painting techniques date back to the early days of filmmaking, the digital era has usher in new methods, with artificial intelligence (AI) playing an important role. Matte painting, at its core, is about crafting the illusion of a realistic environment that may not exist in reality. Matte painting is a technique used to complement the overall look and concept of a film, whether it’s depicting the interior of a star freighter or the smoky backdrop of an industrial city. Traditional painting techniques have been replaced by digital compositing in the digital age, and AI technology has further streamlined the process. While tools like Wacom tablets and Photoshop are still useful, AI provides a more efficient way to build the foundations of these conceptual worlds. The key is to recognise AI as a tool driven by intent. If the use of AI helps to tell a compelling story or create meaningful art, it should be welcomed rather than feared. AI is here to stay, evolving and becoming more sophisticated all the time, and there’s a lot of room for collaboration between human creativity and artificial intelligence.

 

AI integration in the creative sector presents significant challenges, including ethical concerns, bias, and the complex relationship between AI and human creativity. While AI can be intelligent, it is unaware of larger contexts and emotions. UNESCO highlights the impact of AI on cultural aspects, education, and creativity, as well as potential information access disparities. In the realm of AI-generated content, authorship issues as well as concerns about piracy, originality, and unintended exploitation arise. The ease with which convincing fake content can be created raises ethical concerns, which are mitigated by ongoing developments in AI-based methods to detect such content. Algorithmic bias, caused by unevenly distributed training data, is a major concern, necessitating continuous monitoring and periodic retraining of AI systems. There are several ethical classifications that focus on design, analysis, and codes to ensure ethical AI use. AI, such as Instagram’s anti-bullying AI, can be used to identify and resolve ethical issues. In creative processes, AI serves as a tool for tasks that are beyond the capabilities of humans, but human oversight is still essential. Current supervised learning models struggle with unstructured creative outputs, necessitating refinement, particularly in generative models such as GANs. Addressing ethical concerns, mitigating biases, and acknowledging the collaborative nature of AI and human creativity in the creative sector are all required for responsible AI integration. (Anantrasirichai and Bull, 2022).

REFERENCE: John Lewis the snowman’s journey Christmas Advert

 

Can you make 4 x AI garneted shots for this advert:

 

 

Prompt: Mountain and tress Covered in snow, with two big trees on Extreme right and Left side of Screen, Keep Middle space Empty for the subject.

I tried Multiple Times, but I was Not getting the result I want. So, I use Different AI tools to generate Background Like first I tried in Deep.AI used same Prompt.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Then used another AI Tool perchance to get what I want it was not same but much closer to what I want.

 

 

 

 

 

 

 

 

 

 

 

Prompt: Tress Covered in snow, with two big trees on Extreme right and Left side of Screen, Keep Middle space Empty for the subject.

 

 

 

 

 

 

 

 

 

Then I use Prompt Remove to remove unwanted tress then copy and pasted same image and taken another tree from the image. So, AI is just algorithms its not creative it required Human touch to achieve the brief I want.

Then I created rock fence using prompt in photoshop: A rock fence or boundary covered with snow, extend this rocks boundary barrier covered with snow.

Then I added Snowman using simple prompt: Two Snowman Looking at each other.

Then I added colour, grain, and blur to look close to realistic, but I know that two snow man are cartoony. But I also get many errors and faults when generating using AI for example:

So, I have to remove the unwanted part of image that was generated so there is always some correction and modification.

Finally, while AI’s role in the creative sector is growing, issues of ethics, data bias, and social impact highlight the need for responsible development. Despite advancements, AI in creativity remains human-centric, with the goal of augmenting rather than replacing human creativity.

AI-powered visual effects integrate seamlessly into workflows, acting as collaborative tools that enhance rather than disrupt creativity. AI becomes a harmonious companion in the creative process by recognising and complementing individual artists’ styles. It contributes to a collaborative dance between technology and human creativity by identifying gaps in narratives and suggesting plot twists. Ethical considerations include preserving artistic diversity in the face of automation and ensuring AI empowers rather than replaces artists. Maintaining artistic control and addressing ethical concerns about deepfakes, skill gaps, data privacy, and cultural sensitivity are among the challenges. To overcome these challenges, human-AI collaboration, transparency, ethical guidelines, ongoing education, and potential regulations are required. As animation studios illuminate the canvas of innovation, the horizon of AI-driven visual effects holds limitless possibilities, from drawing inspiration from art history to crafting biodiverse landscapes. Finally, the role of AI in animation represents a fusion of technological marvel and humanity’s pursuit of creativity, promising a transformative era in which storytelling is redefined through the synergy of AI and artistic expression. (Ai-driven visual effects, 2023).

 

Conclusion

The incorporation of generative AI within Photoshop has resulted in a seamless union of manual artistry and technological innovation, demonstrating the potential for exceptional results when human skill is combined with AI capabilities. As an artist, I regard manual matte painting as a difficult but immensely rewarding process that necessitates a distinct touch and narrative talent. While generative AI improves efficiency and accessibility, it cannot replicate the rich intricacies and emotional nuances of human-created matte paintings. The combination of manual artistry and generative AI, in which AI augments rather than replaces the creative process, holds the promise of extraordinary results. The key to reaching new heights in matte painting is to embrace the interaction of manual artistry and AI advancements. When considering this artistic journey, the interaction between classical matte painting and generative AI resonates as a symphony in the grand climax of artistic progress. Each note in its own rhythm reflects the essence of creativity, with the ending not serving as a final chord but as a bridge to the next movement in the orchestration of art and technology. beautifully expresses how the craftsmanship inherent in traditional matte painting becomes a nostalgic melody—a tribute to the human touch in which each stroke tells a story and every hue transmits an emotion. The pixelated revolution of generative AI introduces a futuristic rhythm to this ageless dance. Neural rendering approaches and AI-powered virtual sets herald a new era of efficiency—a digital sonnet written in algorithmic language. As the sounds of tradition blend with the beats of innovation, a convergence of contrasts emerges. Gomez and Wang’s (2019) discussion of AI-generated landscapes connects the artist’s difficult journey to the algorithmic efficiency of artificial intelligence. It’s a duet in which human imagination meets digital precision, yielding a symphony of possibilities. The symphonic interplay between classical matte painting and generative AI unfolds as a testament to the evolving landscape of art in this narrative. As artists navigate the intricate balance between tradition and innovation, each stroke on the canvas becomes a dialogue between the tangible history of manual artistry and the boundless potential of AI-driven advancements. This harmonious collaboration resonates not only as a celebration of artistic diversity but also as an exploration of the symbiotic relationship between human ingenuity and the ever-evolving realm of artificial intelligence in the canvas of creativity.

 

 

REFRENSES:

Ai-driven visual effects (no date) HOUND STUDIO. Available at: https://hound-studio.com/blog/ai-driven-visual-effects/ (Accessed: 11 January 2024).

Anantrasirichai, N. and Bull, D. (2022) ‘Artificial intelligence in the creative industries: a review’, Artificial Intelligence Review, 55(1), pp. 589–656. Available at: https://doi.org/10.1007/s10462-021-10039-7.

Artificial intelligence and film industy (no date) Kate Xagoraris. Available at: https://www.katexagoraris.com/artifical-intelligence-and-film-ind (Accessed: 14 January 2024).

Bolter, J.D. (2016) The next step: exponential life. Translated by W. Matthews. [Bilbao, Spain]: BBVA.

Expiredinvalid (no date). Available at: https://r1.vlereader.com/ErrorInvalidExpired (Accessed: 11 January 2024).

Hageback, N. and Hedblom, D. (2021) Ai for arts. 1st edn. CRC Press. Available at: https://www.perlego.com/book/2554799/ai-for-arts-pdf (Accessed: 11 January 2024).

Mántaras, R.L. de (no date) ‘Artificial intelligence and the arts: toward computational creativity’, OpenMind. Available at: https://www.bbvaopenmind.com/en/articles/artificial-intelligence-and-the-arts-toward-computational-creativity/ (Accessed: 11 January 2024).

Manovich, L. (2023) ‘Chapter 5 – AI images and Generative Media.’, in Artificial Aesthetics: A Critical Guide to AI, Media and Design. Available at: http://manovich.net/content/04-projects/168-artificial-aesthetics/lev-manovich-ai-aesthetics-chapter-5.pdf (Accessed: 30 October 2023).

Miller, A.I. (2019) The artist in the machine: the world of AI powered creativity. Cambridge, Massachusetts: The MIT Press.

O’brien, M. and Lajka, A. (no date) AI tools can create new images, but who is the real artist? Available at: https://techxplore.com/news/2023-01-ai-tools-images-real-artist.html (Accessed: 11 January 2024).

pholmesphd (2021a) ‘History of matte painting’, Peter Holmes PhD Journal, 19 December. Available at: https://pholmesphd.home.blog/2021/12/19/matte-painting/ (Accessed: 14 January 2024).

Sancto, C.D.M. (2020) ‘Review: ai art: machine visions and warped dreams , by joanna zylinska’, Afterimage, 47(4), pp. 82–86. Available at: https://doi.org/10.1525/aft.2020.47.4.82.

Saunders, F. (2023) ‘What AI means for the art’, The Daily Telegraph, 27 May, p. 4.

The dawn of generative ai: a threat to creatives or a boon? (2021). Available at: https://www.wowmakers.com/blog/generative-ai/ (Accessed: 11 January 2024).

Using impossible software to create impossible worlds – the rise of AI and the art of digital painting. | Dazzle Pictures (no date). Available at: https://www.dazzle.pictures/blog/using-impossible-software-to-create-impossible-worlds (Accessed: 14 January 2024).