This is the latest entry in our ‘State of the art-ful AI’ series, where we upload a photo to Chat GPT 4 and ask it for a photo description, then ask DALL-E 3 to generate an image from the description. How similar will they be?
Why this workflow?
Less than a year after its launch, ChatGPT and DALL-E have already seen multiple iterations. In October 2023, ChatGPT 4 gained the capability to describe uploaded photos. To track the progress of generative AI, we uploaded these AI descriptions to the Dall-E 3 image generation model to see how faithful its output is to the original photo. A variety of photos have been selected which we’ll bring you throughout November and December.
Our original photo
This photo was taken in Bristol, England. It depicts the Bristol planetarium as part of the complex that is now called ‘Wethecurious‘.
The photo as described by Chat GPT 4 in 150 words
This image captures a busy public square on a clear day with a mix of clouds and blue skies. At the center is an impressive, spherical structure with a highly reflective metallic surface, resembling a giant mirror ball. The sphere’s polished exterior creates a distorted panorama of the surrounding area, including the buildings and the square itself. The architecture around the sphere is modern, with a mix of plain and reflective facades.
People are going about their day in the square, with activities suggesting a relaxed, leisurely atmosphere. There are adults walking, some pushing strollers, while children are playing. Everyone is dressed in casual, comfortable clothing, indicating a cool but pleasant weather.
The reflection on the sphere is crisp and clear, capturing inverted images of the square and sky, which adds a dynamic and surreal quality to the scene. The design of the sphere is such that it stands out as a landmark, likely to draw the attention and curiosity of passersby. The square appears to be a shared space that caters to both pedestrians and cyclists, promoting a community vibe in an urban setting.
Photos generated by Dall-E 3 using the AI description
Spot the difference
Using a separate chat, we asked Chat GPT what differences it finds between the original photo and the FIRST image generated above:
Here’s a comparison of the two images using a numbered list:
- Location Context: The first image shows a more academic or institutional background with simpler buildings, while the second image is set in a bustling urban square surrounded by modern architecture.
- People: The people in the first image seem to be going about their day casually, with some children in sight, whereas the second image has a denser crowd with more diverse activities, including people on bicycles.
- Weather and Sky: The sky in the first image has a few clouds with a clear blue sky, suggesting a calm day. In contrast, the second image has a dramatic sky with more pronounced cloud formations and a more dynamic lighting, possibly indicating a different time of day or weather conditions.
- Reflections: Both images feature a large reflective sphere, but the reflections differ. The first image has a clear reflection of the immediate surroundings, while the second image’s sphere reflects the sky and an inverted image of the sphere itself, adding a surreal touch to the scene.
- Sphere Design: The sphere in the first image appears to be resting on a base that is part of its structure, giving it a half-sphere appearance. The second image’s sphere is a full sphere with a small, detached base, which makes it look as if it’s floating above the ground.
- The Square: The first image shows a more confined space with a limited walking area around the sphere, while the second image depicts a vast, open square with ample space for pedestrians.
- Lighting and Tone: The lighting in the first image seems natural and evenly distributed, whereas the second image has a warmer tone with more contrast and emphasis on sunlight reflecting off the sphere and the buildings.
- Artistic Style: The first image is a photograph with realistic proportions and colors, and the second image, while also a photo-realistic representation, has a touch of an artistic rendering, possibly due to editing or filters that enhance its colors and contrast.
Example workflow video
Here’s our workflow in action, using an ice cream truck image. This is not an explainer video – It shows the main steps of the above ChatGPT 4 workflow in November 2023.
