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 using 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 image shows one of the many ‘Shaun the sheep’ displayed across Bristol, England as part of the Shaun the Sheep trail in 2015
The photo as described by Chat GPT 4 in 150 words
This image features a large, colorful sculpture of a sheep, standing upright on a rectangular green platform, which appears to be a part of an outdoor exhibition or event. The sculpture is whimsically painted with a variety of bold, solid colors and patterns. The head of the sheep is green on the top, blending into a black and white checkered pattern on the face, with large cartoonish eyes and red around the nose area, giving it an endearing expression.
The ears are painted blue and yellow, and the body showcases a mix of vibrant red and a pattern that includes blue, yellow, green, and red stripes. The legs are also patterned, with one featuring vertical stripes and the others with different solid colors. In front of the sculpture is a sign that says “PLEASE STAND BY” along with logos, suggesting it might be part of a series or trail, which is common in public art installations. The background shows a pedestrian area with people walking by, market stalls with red and blue canopies, and a city environment, suggesting the setting is an urban square or street.
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:
- Color Scheme: The first sheep has a multicolored design with each body part featuring different patterns and colors, such as a green head, black and white checkered torso, red face, and multicolored limbs. The second sheep has a black body with a green patchwork design and rainbow-striped legs.
- Patterns: The first sheep’s patterns are more geometric and include checkers, stripes, and solid colors. The second sheep has a consistent patchwork pattern resembling a quilt or stained glass, with the rainbow stripes at the legs being the only deviation.
- Head: The first sheep has a green head with what appears to be a partial clear section on the top, whereas the second sheep’s head is entirely black with green patchwork.
- Face: The first sheep has a red area around its eyes, a white mouth area, and a clear section over the snout. The second sheep has a heart-shaped nose, green patchwork around the eyes, and no clear section.
- Legs: The legs of the first sheep are multicolored with different patterns on each leg, including stripes and solid colors. The second sheep’s legs feature a consistent rainbow-striped pattern.
- Eyes: Both sculptures have wide, white eyes with black pupils, but the first sheep has its eyes pointing outward, and the second sheep has its eyes pointing forward with a touch of red around them.
- Ears: The first sheep has one blue ear and one multicolored ear, while the second sheep has both ears black with a green patchwork design.
- Tail: The first sheep’s tail is blue, while the second sheep’s tail matches the body with a black and green patchwork design.
- Background and Setting: The first image shows a city street background with tents and people, indicating an urban setting, while the second image has a more staged appearance with market stalls and seating areas in the background.
- Base: Both sheep stand on a green base, but the first one has plain green edges, and the second one has a black border with a QR code and text plate on it.
These artworks are likely part of a public art trail featuring Shaun the Sheep sculptures, each uniquely decorated by different artists.
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.
