ComfyUI Node: FlashFace Sampler

FlashFace Sampler:

FlashFace_Sampler_NuA is a specialized node designed to generate high-quality, ultra-detailed facial images using advanced AI models. This node leverages the power of FlashFace models to create photographic and UHD 4K quality images, ensuring that the generated faces are realistic and visually appealing. The primary goal of this node is to provide AI artists with a tool that can produce detailed and accurate facial representations, which can be used in various creative projects. By utilizing reference faces and fine-tuning various parameters, you can achieve precise control over the output, making it an invaluable asset for generating professional-grade facial imagery.

FlashFace Sampler Input Parameters:

flashface_models

This parameter specifies the FlashFace models to be used for generating the facial images. It is a required input and ensures that the node uses the appropriate model for the task.

reference_faces

This parameter allows you to input reference images that the node will use as a guide to generate the new facial images. It is a required input and helps in achieving a closer resemblance to the desired output.

positive

This is a string parameter where you can input positive prompts to guide the image generation process. The default value is "best quality, masterpiece, ultra-detailed, UHD 4K, photographic". These prompts help in enhancing the quality and details of the generated images.

negative

This is a string parameter where you can input negative prompts to avoid certain undesirable features in the generated images. The default value is "blurry, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, bad anatomy, watermark, signature, cut off, low contrast, underexposed, overexposed, bad art, beginner, amateur, distorted face". These prompts help in filtering out unwanted characteristics.

steps

This integer parameter defines the number of steps for the image generation process. The default value is 35. Increasing the number of steps can improve the quality of the generated image but may also increase the processing time.

height

This integer parameter sets the height of the generated image. The default value is 768, with a minimum of 256 and a maximum of 4096. Adjusting this parameter allows you to control the resolution of the output image.

width

This integer parameter sets the width of the generated image. The default value is 768, with a minimum of 256 and a maximum of 4096. Adjusting this parameter allows you to control the resolution of the output image.

face_bbox_x1

This float parameter defines the x-coordinate of the top-left corner of the bounding box for the face. The default value is 0.4, with a range from 0.0 to 1.0 and a step of 0.1. This helps in specifying the region of interest for the face in the image.

face_bbox_y1

This float parameter defines the y-coordinate of the top-left corner of the bounding box for the face. The default value is 0.3, with a range from 0.0 to 1.0 and a step of 0.1. This helps in specifying the region of interest for the face in the image.

face_bbox_x2

This float parameter defines the x-coordinate of the bottom-right corner of the bounding box for the face. The default value is 0.6, with a range from 0.0 to 1.0 and a step of 0.1. This helps in specifying the region of interest for the face in the image.

face_bbox_y2

This float parameter defines the y-coordinate of the bottom-right corner of the bounding box for the face. The default value is 0.6, with a range from 0.0 to 1.0 and a step of 0.1. This helps in specifying the region of interest for the face in the image.

lamda_feat

This float parameter controls the weight of the feature loss during the image generation process. The default value is 1.2, with a step of 0.1. Adjusting this parameter can influence the balance between different features in the generated image.

lamda_feat_before_ref_guidence

This float parameter controls the weight of the feature loss before applying reference guidance. The default value is 0.85, with a step of 0.1. This parameter helps in fine-tuning the initial stages of the image generation process.

face_guidence

This float parameter controls the strength of the face guidance during the image generation process. The default value is 3.2, with a step of 0.1. Increasing this value can enhance the accuracy of the facial features in the generated image.

step_to_launch_face_guidence

This integer parameter defines the step at which face guidance should be launched during the image generation process. The default value is 750. This parameter helps in timing the application of face guidance for optimal results.

text_control_scale

This float parameter controls the scale of text-based control during the image generation process. The default value is 7.5, with a step of 0.1. Adjusting this parameter can influence the impact of text prompts on the generated image.

seed

This integer parameter sets the seed for random number generation, ensuring reproducibility of the results. The default value is 0, with a range from 0 to 0xffffffff. Using the same seed will produce the same output for the same input parameters.

FlashFace Sampler Output Parameters:

IMAGE

The output of this node is an image generated based on the input parameters and reference faces. The generated image will be a high-quality, ultra-detailed facial representation that adheres to the specified prompts and settings. This output can be used in various creative projects, providing a realistic and visually appealing facial image.

FlashFace Sampler Usage Tips:

• To achieve the best results, use high-quality reference faces that closely match the desired output.
• Experiment with different positive and negative prompts to fine-tune the characteristics of the generated image.
• Adjust the steps parameter to balance between image quality and processing time.
• Use the face bounding box parameters to focus on specific regions of the face for more accurate results.

FlashFace Sampler Common Errors and Solutions:

"Model not found"

• Explanation: The specified FlashFace model could not be located.
• Solution: Ensure that the correct model name is provided and that the model is available in the specified directory.

"Invalid reference face"

• Explanation: The provided reference face image is not valid or cannot be processed.
• Solution: Check the reference face image for any issues and ensure it is in a supported format.

"Parameter out of range"

• Explanation: One or more input parameters are outside the allowed range.
• Solution: Verify the values of all input parameters and ensure they fall within the specified ranges.

"Seed value error"

• Explanation: The seed value provided is not within the acceptable range.
• Solution: Ensure the seed value is between 0 and 0xffffffff.

"Image generation failed"

• Explanation: The image generation process encountered an error.
• Solution: Review all input parameters and reference images, and try adjusting the settings to resolve any potential conflicts.