Monty Final Project

New GANT chart

My new plans to create one single film shot entirely in a studio with green screens and projectors has totally changed my schedule. I am planning to shoot 11-16th April and will spend the coming weeks producing the shoot, doing test shoots, and working on the CGI world building. This will mean that by the shoot date I will be totally prepared and will have a good idea what the world will look like, allowing me to light the actors and block the scenes accordingly. I will also spend time researching both artistic and academic resources to ensure I am approaching the film with as much information as possible.

I also think it’s maybe a good idea to shoot the first part of the film before the main shoot. This will make the main shoot far less stressful and help discover if we have any problems before it is too late

pre-production Gant Chart
I also made this preproduction Gantt chart to keep me on track with all the different tasks needed to be completed before the shoot

TO DO

Update my Gantt chart
Write script
Cast actors (started)
Research films for visual style
Research Academic context
Work on CGI world building (started)
Test shoots
Produce film

I have made some major changes to my idea and how I will execute my project.

Firstly, all the filming will be done in a studio using green screen, set building, and projectors to place the actors inside the environments. I have much experience filming and editing but I have never before shot dialogue. As such, I was over optimistic about how easy it would be to shoot dialogue outside, but after some thought about how this would actually be achieved, I began to realise it would become very stressful and most likely wouldn’t work. As such, it is now going to be a studio shoot. This also means I have far more control over environments, and I intend to use several AI-postproduction techniques to create the world the story exists in. These changes in my approach Will make the project more feasible to complete and give me far greater control as I will not be relying on good weather and all the other problems that can come with shooting outdoors.

Secondly, instead of making it three separate films I will make one film with a narrative that encompasses the stories from the three films. This means I will use the same characters, actors, and visual styles and the narrative well flow.


TO DO

Update my Gantt chart
Write script
Cast actors
Research films for visual style
Research Academic context
Work on CGI world building

I trained a style Gan on photographs I had collected with my image scraper of nature, cities, and explosions. This stylegan then creates synthetic amalgamations of the patterns between the images, forming at local minimums. This means there are places in this style gan’s latent space where are the images are entirely buildings, or entirely nature, and others where they are a mixture of the two. This means I could select images from a range of nature to city, and then and export a film of the model morphing between these images. The result is to create a loopable film that documents the rise of cities out of nature and the inevitable and violent full of the cities back into the ground and the forests growing back up around then.

For the second film, I imported the film into blender where are used texture notes to separate the bright rbg values and make the emmit light, as well as adding a bloom effect. This makes the lights and fire appear real and three-dimensional

must
An AI generated short film that sets the context and themes of the narrative films.
A short film that explores the dangers and the power of artificial intelligence.

Should
A short film that establishes characters and the world.

Could
Another short film that further establishes the characters and the stakes.

Won’t
Shoot anything out side

Specific
Create a film, compromised of three parts and an opening sequence, that explores the strength and potential dangers of artificial intelligence and the power of the unrefined human spirit.

Measurable
Completion of scripts, show filming, and final cuts that successfully capture the relevant themes and ideas. Most important is the quality of the final film some selves and there success in capturing the themes.

Actionable
write scripts, produce the films, film the films, edit, sound design, and colour grade the films. write scripts, produce the films, film the films, edit, sound design, and colour grade the films. Each of the steps for each film is its own action and requires its own care and consideration.

Relevant
The most important parts of this project are the opening sequence, which uses AI tools to give context to the films, and the final film, which will most directly explore the themes of artificial intelligence. As such, I will begin working on the opening sequence first, and should I find the first two films taking too long, I will put them aside and focus on the final film.

Also relevant, is the artistic quality of the finished films. They are pieces of art that should be charged above all else own artistic merit.

Timebound
This is largely covered in my Gantt chart, but in summary, I plan on shooting the first two films and making the opening sequence in the first six weeks, leaving me the second six weeks to shoot the final film. I will also focus my research and thesis work in the first half of the project.

This is my gant chart for the project. What I have set out is very ambitious and I am prepared to strike the second film if I am unable to meet the deadlines. I will be realistic about this but I intend to try.

As discussed in mind meeting with Anna, I am planning to make an opening sequence using AI generated images along with AI narration to establish the context of the films.

The purpose of this film Will be to immerse the viewer within the world. To do this, I plan to train a stylegan of images of both cities and nature and then make a latent walk film that slowly progresses from scenes of nature to colossal cityscapes. I then plan I’m creating an AI depth map for the film which I will use to displace the film, which can be used to create hard shadows. This rise of concrete and glass out from the natural world until no more nature can be seen Will establish that this new world is in entirely artificial. When the audience then sees the characters set within natural worlds, it will raise the question of wether the events are taking place within and imagined world.

To begin with, I modified an image scraper script I have written with the Bing API and python to download the images I will use to train the StyleGan. Given a list of search terms, the scraper worked until completed, saving me the incredibly boring task of downloading the images myself. The scraped images are usually relevant, although many irrelevant images also are downloaded, for example the search ‘jungle sunset’ also returns many images of people in the jungle, which will only confuse the stylegan

The images I have scraped vary widly in subject. Although the stylegan will struggle with this, I think the results may be extraordinary

My next steps are:

Film a test shoot for The Briefcase. This sets for this film Will be a table and chairs in various outdoors Locations; the beach, forests, and moar. It is necessary to test this to find the best time of day to shoot and how best to light the scenes.

Find locations for The Briefcase

Find actors for both films

Film the ocean for Waves of Sorrow. I need long video this of waves crashing which I can project onto the set walls.

My meeting with Anna has been very helpful in clarifying my ideas and forward direction.
The key takeaways were:

My research should include Visual references to inform the style I will film in as well as films and media that cover similar seems to mine, and also Academic research into the topic of artificial intelligence to inform the context of my films.

Because my films will focus around artificial intelligence, I should use some AI tools in the creation of my films. I was sceptical about this at first, as I did not want to cede any creative control to an algorithm, for example by allowing chat GTP to write some of my script. However, after some conversation we had the idea of allowing AI to create the opening sequence of the films with the instructions to give context to the narratives. In doing this, I am not relinquishing artistic control to the algorithms but instead I’m using them as a co-collaborator. The effect of opening the films on an AI created sequence Will be to immerse the viewer within the context of AI, helping them to understand the events that will later transpired.

Something I have been worrying about has been the scale of the project, as three films is a lot to complete in as many months. It is important to me to create the three films as each one is essential in the wider narrative. However, Anna suggested the idea that it is not necessary for me to make all three films before my graduation and that I can make one or even two of them after this project is completed. This will mean I am able to focus on one or two films and make them well rather than rushing through all three. This being said, I still intend to try to make or three but it is good to know I have the option to fall back on.