#DALL-E Fails to Depict Characters Running Toward the Camera Despite Explicit Prompts

1 messages · Page 1 of 1 (latest)

scenic joltBOT
#

Reported by @marble gust

Bug Report: DALL-E Fails to Depict Characters Running Toward the Camera Despite Explicit Prompts
`Steps to Reproduce`

Use a prompt specifying a character running toward the camera, such as:
"The camera is positioned directly in front of the teen, capturing her face and torso fully visible as she runs toward the viewer. Her back and legs are not visible."
Ensure the prompt includes detailed perspective instructions and motion context.
Use variations of terms like "fleeing," "escaping," or "running toward" to eliminate ambiguity.
Generate the image and review the results.

`Expected Result`

The generated image shows the character running toward the viewer, with her face and torso fully visible in a frontal view. The zombies should be in the background, clearly pursuing her.

`Actual Result`

The generated images consistently depict the character running away from the camera, with rear-facing or side-facing perspectives, despite the prompt explicitly emphasizing a frontal view.

`Environment`

Platform: DALL-E (via ChatGPT)

#
Additional Information

Please provide relevant details to help resolve the issue, such as:

  • ChatGPT Shared Link (if applicable).
  • Screenshots or videos demonstrating the problem.

-# ➜ Need to contact support? Visit the OpenAI Help Center.

marble gust