First experience with Bing’s Image Creator.

18

October

2023

No ratings yet.

This week I tried the AI-generated image tool for the first time… and it did not disappoint!

I chose Bing’s Image Creator for two reasons. One, too many students already wrote their blogs on Dall-E, and second, Bing’s Image Creator is free. Let’s see what the fuss is about with that out of the way. I am going to find out the difference between Image Creator and Google Images, and if one is better than the other.

1st Prompt: a space cowboy running on the moon with a monkey on his shoulder

From Google Images, I got the following.

From Bing’s Image Creator, I got the following.

2nd Prompt: most beautiful place on earth

From Google Images, I got the following.

From Bing’s Image Creator, I got the following.

I think the main difference between these two is the customization from Image Creator. With that tool, I can give more specific prompts and Image Creator will be able to accurately generate the image that I want. On the other hand, for simpler prompts (like the second one), there is a closer similarity with the results from Google Images and Bing’s Image Creator. This made me realize that for general prompts, Google Images is more than enough for anyone to use. At the same time, for anything more specific and unique, one can go the extra mile to use AI-generated images to get the images they want. Regardless, this has been an insightful experience and will help me use the tool better in the future.

References:

Please rate this

Why does AI struggle to create images of human hands?

18

October

2023

No ratings yet.

I was really surprised to learn during one of our Information Strategy lectures that AI generally struggles with generating pictures of human hands. First, I wanted to try that out and second, I wanted to know why.

I used two AI tools to test this anomaly, Bing Image Creator and ChatGPT-4 with the DALL·E 3 beta. I used the following prompt: “Create a realistic-looking picture of a professor in a university lecture hall. The professor is standing in front of the class and is holding a presenter in the one hand, and a coffee cup in the other hand.” At first sight, you don’t really see the issue but when you look more closely, you see how weird some hands look. In picture 1, the woman is missing a finger of her right hand, and her left hand is looking especially unnatural. The professor’s presenter in picture 2 is floating in the air above his hand and the woman in picture 3 appears to have three hands, two of which are holding the coffee cup and another one is holding the presenter. Overall, all these generated images look very good and natural in my opinion, except for the hands.

But why is generating images of human hands such a problem for AI? Firstly, the models are 2D image creators that do not understand the three-dimensional nature of a human hand. (Hughes, 2023). Secondly, their training data mostly focused on other parts of the human body, such as the face (Hughes, 2023). Therefore, the AI tools have especially big problems with creating images of hands if you provide a context in which the hands must appear, such as holding specific objects in my case.

I’m curious to see how this topic evolves in the future and how long it will take AI tools to get better at generating human hands. I look forward to your comments!

References

Hughes, A. (2023, February 4). Why AI-generated hands are the stuff of nightmares, explained by a scientist. BBC Science Focus. https://www.sciencefocus.com/future-technology/why-ai-generated-hands-are-the-stuff-of-nightmares-explained-by-a-scientist

Please rate this

The prospective role of AI in cinematography

29

September

2023

No ratings yet.

Earlier this week, I went to a movie screening at the cinema that was made in honor of one of the most celebrated Directors of Photography in the Netherlands, Robby Müller. This experience has made me ruminate on many aspects of the world of cinema, and more specifically how so many jobs in cinema are not appreciated enough. Therefore, this blog post will focus on the topic of Director of Photography, which is the starting point of my discovery of the hidden jobs in cinema.

So what is a Director of Photography? What do they exactly do? Directors of Photography play a paramount role in the making of movies as they help the directors to shoot movies with an artistic touch, through the different angle shots that they would propose. They assist in managing the right lighting and translate the vision of the director into the shooting, ultimately creating an idiosyncratic aesthetic for the movie (ScreenSkills, n.d.).

However, with the advent of Artificial Intelligence (AI), the process of movie-making might be disrupted (Smith, 2023). Throughout history, the world of cinematography has been shaken by technological advancements, from the integration of audio into movies to the introduction of color imaging and computer-generated imagery (Bernardo, 2016). AI would not make an exception to that trend. Consequently, it is easy to imagine how cinema and more specifically Directors of Photography could directly get some of their tasks delegated to AI. This could include things like storyboarding and color grading.

To explore this concept further, I decided to experiment with image generation myself with Bing AI — which is powered by DALL-E for image creation (Kan, 2023). I inserted 3 prompts by pretending to play the role of a movie director and asking to generate pictures for creative inspiration.

The first prompt was: “Present a mysterious journey into the jungle with few soldiers in a boat witnessing the absurdity and horrors of the war. Show their shocked faces and their concerns through an unorthodox angle shot so that it leaves a mark on the audience.”; Bing AI gave me the following pictures:

The second prompt inserted was: “Generate an image of an old antagonist that transpires philosophical energy. Present him through an angle that would give chills to the audience.”; Here are the images generated:

The last prompt I suggested was: “Try to represent the magnum opus of Pablo Picasso “Guernica” in a real-life setting. Make the scene as lugubrious and tragical as possible.”; The AI gave me the following:

Looking at those results, Bing AI was able to translate the visions in a more or less clear manner. As can be seen with the results of the first and second prompts, the faces and body features are disfigured but the energy that transpires is astonishing. However, regarding the last prompt, due to the abstract nature of the initial inquiry, with Picasso’s painting of “Guernica”, the result came out accordingly abstract. Thus, Bing AI is capable of creating an overall coherent image content, but when scrutinizing, it is observed that the AI misses on the specific details of the subject it portrays. Moreover, as seen, the clearer the prompt, the better the results.

All-in-all, AI tools already show great potential as idea generators and storytelling aids. I believe that with further progress, this technology could complement and enhance the human creativity of the Director of Photography and push the boundaries of visual storytelling in an unprecedented way.

Bibliography

Bernardo, A. B. (2016, March 10). 7 Advances in Technology that have Revolutionized the Film Industry | OpenMind. OpenMind. https://www.bbvaopenmind.com/en/technology/innovation/7-advances-in-technology-that-have-revolutionized-the-film-industry/

Kan, M. (2023, April 7). With DALL-E integration, AI-Powered Bing can also generate images. PCMAG. https://www.pcmag.com/news/with-dall-e-integration-ai-powered-bing-can-also-generate-images

ScreenSkills. (n.d.). Director of photography (DoP). ScreenSkills. https://www.screenskills.com/job-profiles/browse/film-and-tv-drama/technical/director-of-photography-dop/

Smith, D. (2023, March 23). ‘Of course it’s disturbing’: will AI change Hollywood forever? The Guardian. https://www.theguardian.com/film/2023/mar/23/ai-change-hollywood-film-industry-concern

Please rate this