AI Prompts for Ensuring Consistent Subject Identity

Run a baseline test prompt using only the core subject description: '[PLACEHOLDER] standing in a park'. Analyze the output to determine if the subject retains its identity and characteristics without any additional context or details. Provide specific observations on how the lack of context affects the subject's representation.

Test the following reference prompt: 'A [PLACEHOLDER] in a sunny park, smiling'. Analyze the output to determine if the subject remains consistent throughout the clip, and identify any discrepancies in the subject's representation within this familiar context.

Change the style parameter of the video to a different setting, such as 'CINEMATIC' or 'DOCUMENTARY'. Analyze how this change affects the consistency of the subject throughout the clip and provide specific observations on any differences noted in subject representation.

Generate four variations of the prompt: '[PLACEHOLDER] walking through a bustling city'. Analyze each variation to determine if there is a consistent subject identity maintained throughout the clips, and provide a brief explanation of your findings for each variation.

Generate a video clip featuring '[PLACEHOLDER] in a park, wearing a blue shirt'. After generating the clip, compare it against a reference image of the desired subject and identify any discrepancies in appearance, noting specific differences in color, style, and overall likeness.

Rephrase the following prompt to include specific descriptors: '[PLACEHOLDER]'. Ensure to add details such as clothing color, hair color, and location, for example, 'wearing a blue shirt, with brown hair, standing in a park'. This added specificity will help maintain consistency in the description.

Incorporate negative prompts into your description to refine the output. For example, describe a scene as 'A [PLACEHOLDER] in a park, not wearing a hat' to specifically exclude unwanted variations. Provide three additional examples of negative prompts that could enhance the clarity of the description.

Switch to a different model version in the settings and evaluate the output for consistency with your subject. Provide a comparison of the results from the previous model version and the new one, highlighting any improvements or issues observed.

Restructure the following prompt to place the subject first: '[PLACEHOLDER] in a park, cinematic style'. Explain how this change emphasizes the subject's identity and suggest any additional modifications to enhance the overall impact of the prompt.

In your video project, ensure that the lighting remains consistent throughout the clip. For example, specify the lighting conditions as '[PLACEHOLDER] in a park, during golden hour' to create a unified appearance for the subject. Explain how this consistency can enhance the overall visual quality of the video.

Generate four new variations based on the prompt: '[PLACEHOLDER] walking through a park'. Ensure that the subject identity remains consistent in each variation, and provide a brief analysis of any changes in the subject's representation across the outputs.

Compare the new prompt output against the Stage 1 baseline output. Identify and describe any discrepancies in the subject's appearance, focusing on changes in attributes such as color, size, and clarity. Provide specific examples of these differences and suggest potential reasons for any observed variations.

Test the AI's ability to maintain identity recognition under challenging conditions by using the following complex subject description: '[PLACEHOLDER], a young woman with glasses, reading a book in a park'. Analyze the results and provide feedback on any difficulties the model encounters in accurately identifying the subject throughout the clip.

Lock the seed in the settings to ensure consistency in the generated outputs. Then, regenerate the prompt using the following template: '[PLACEHOLDER] in a park'. Provide the results and compare them to verify the resolution consistency across generations.

Check the video with the subject changing mid-clip by running it in different aspect ratios. Start with 16:9 and analyze whether the subject remains consistent throughout the clip. Provide a detailed observation of any changes in the subject's visibility or framing as the aspect ratio is adjusted.

Create a reusable base template for maintaining consistent subject identity in VEO. The template should follow this structure: '[PLACEHOLDER], wearing [CLOTHING], with [FEATURES], in [SETTING]'. Ensure that this template is saved for future use to prevent subject changes mid-clip.

Create a negative prompt bank that includes phrases such as 'not wearing [ITEM]' and 'not in [UNWANTED SETTING]'. Explain how these negative prompts can be utilized to refine your main prompts and ensure consistency in your output. Provide at least five examples of negative prompts that can help clarify desired outcomes.

Develop a workflow for locking seeds in video generation. Specify how to assign a unique seed number for each prompt, ensuring that this seed is consistently applied across all generations to maintain uniformity in the output. Include steps for verifying that the seed is correctly set before generating each video.

Create a detailed checklist for ensuring style consistency in video production. Include specific guidelines for [LIGHTING], [SETTING], and [SUBJECT DESCRIPTORS], and provide examples of how to apply these elements effectively in prompts to maintain a cohesive visual style throughout the video.

Save the following parameter combination template for future prompts: '[PLACEHOLDER], [CLOTHING], in [SETTING], with [LIGHTING]'. Explain how each element can be modified to create diverse scenarios and provide three examples of how to use this template effectively in different contexts.