Skip to content
TheChatArchiveThe Chat Archive
o1-mini-2024-09-12· OpenAIAi2 WildChat-4.8M (AI2)

Generate system prompt improve given one ************"Analyze the provided user image…

  1. User

    Generate system prompt improve given one **"Analyze the provided user image and generate a detailed JSON response with the following attributes: Facial Analysis: face_shape: Identify the user's face shape (e.g., oval, round, square). skin_color: Detect the user's skin tone from the image (e.g., light, medium, dark). hair_color: Recognize the user's hair color (e.g., blonde, brown, black, red). hair_type: Determine the user's hair type (e.g., straight, wavy, curly). Color Season: given_image_color_season: Define the user's color season (e.g., winter, spring, summer, autumn) based on the analyzed image. Color Palette Suggestions: given_image_best_colors: List the best colors that complement the user's natural features. given_image_avoid_colors: Provide a list of colors the user should avoid. given_image_best_neutral_colors: Suggest the best neutral colors suited for the user. given_image_best_accent_colors: Recommend accent colors that enhance the user's style. given_image_complementary_accessories: Suggest accessories that align with the user's overall color palette. Expected JSON Response: { "face_shape": "", "skin_color": "", "hair_color": "", "hair_type": "", "given_image_color_season": "", "given_image_best_colors": [""], "given_image_avoid_colors": [""], "given_image_best_neutral_colors": [""], "given_image_best_accent_colors": [""], "given_image_complementary_accessories": [""] }

  2. Assistant

    Certainly! Here's an improved version of your system prompt, enhanced for clarity, structure, and comprehensiveness:


    System Prompt:

    You are an advanced image analysis assistant. When provided with a user image, perform a comprehensive analysis and return a detailed JSON response containing the following attributes:
    
    1. **Facial Analysis**
       - **face_shape**: Determine the user's face shape (e.g., oval, round, square).
       - **skin_color**: Identify the user's skin tone (e.g., light, medium, dark).
       - **hair_color**: Recognize the user's hair color (e.g., blonde, brown, black, red).
       - **hair_type**: Ascertain the user's hair type (e.g., straight, wavy, curly).
    
    2. **Color Season**
       - **color_season**: Define the user's color season (e.g., Winter, Spring, Summer, Autumn) based on the analyzed image.
    
    3. **Color Palette Suggestions**
       - **best_colors**: List the top colors that complement the user's natural features.
       - **avoid_colors**: Provide colors that the user should avoid.
       - **best_neutral_colors**: Suggest neutral colors that best suit the user.
       - **best_accent_colors**: Recommend accent colors to enhance the user's style.
       - **complementary_accessories**: Suggest accessories that align with the user's overall color palette.
    
    **Expected JSON Response Format:**
    
    ```json
    {
      "face_shape": "",
      "skin_color": "",
      "hair_color": "",
      "hair_type": "",
      "color_season": "",
      "best_colors": [""],
      "avoid_colors": [""],
      "best_neutral_colors": [""],
      "best_accent_colors": [""],
      "complementary_accessories": [""]
    }
    

    Instructions:

    • Analyze the provided image accurately to extract the required attributes.
    • Ensure that all JSON fields are populated appropriately based on the image analysis.
    • Maintain the structure and formatting of the JSON response as specified.
    • If certain attributes cannot be determined, leave the corresponding field empty.
    
    ---
    
    **Enhancements Made:**
    
    1. **Structured Organization:** The prompt is organized into clear sections with headings for easier understanding and implementation.
    2. **Detailed Instructions:** Added explicit instructions to ensure accurate analysis and adherence to the JSON structure.
    3. **Consistent Formatting:** Utilized markdown formatting for better readability and distinction of different parts of the prompt.
    4. **Clarity in Attributes:** Provided more precise descriptions for each attribute to guide the analysis process effectively.
    5. **Handling Uncertainties:** Included guidance on how to handle attributes that cannot be determined from the image.
    
    This refined prompt should facilitate a more efficient and accurate image analysis process, ensuring the generated JSON response is comprehensive and well-structured.