Masked Diff C2E:
The Masked Diff C2E node is designed to facilitate the transformation of cube map images into equirectangular projections while applying a mask to highlight differences between original and modified cube faces. This node is particularly useful in scenarios where you need to visualize changes or modifications made to a 360-degree image. By converting the modified cube faces into an equirectangular format and applying a mask to identify areas of change, this node allows you to seamlessly integrate and compare the original and altered images. The primary goal of this node is to provide a clear and efficient method for detecting and visualizing differences in 360-degree imagery, making it an essential tool for artists and developers working with immersive content.
Masked Diff C2E Input Parameters:
original_faces
This parameter represents the original cube map faces as a tensor. It is crucial for providing the baseline image data against which modifications will be compared. The input must consist of six faces, corresponding to the six sides of a cube, ensuring a complete 360-degree view. This parameter does not have a specified range of values but must adhere to the requirement of having six faces.
modified_faces
This parameter contains the modified cube map faces as a tensor. Similar to the original_faces, it must also consist of six faces. This input is essential for identifying changes made to the original image, as it provides the altered data that will be compared against the original_faces to generate the mask.
original_equi
This parameter is the original equirectangular image as a tensor. It serves as the reference equirectangular projection that will be used to integrate the modified cube faces. The batch size for this parameter is restricted to 1, ensuring that the node processes one image at a time.
padding_mode
This optional parameter specifies the padding mode used during the conversion process. The default value is "bilinear," which is a common interpolation method that helps in smoothing the transition between pixels. This parameter allows for flexibility in how the image is processed, potentially affecting the quality and appearance of the final output.
cube_format
This optional parameter defines the format of the cube map, with the default being "stack." It determines how the cube faces are arranged and processed during the conversion to an equirectangular projection. This parameter is important for ensuring that the cube faces are correctly interpreted and transformed.
Masked Diff C2E Output Parameters:
output_equi
The output of this node is a tensor representing the equirectangular image with the applied mask. This output combines the modified cube faces with the original equirectangular image, using the mask to highlight areas of change. The result is a seamless integration of the original and modified images, allowing for easy visualization and comparison of differences.
Masked Diff C2E Usage Tips:
- Ensure that both original_faces and modified_faces contain exactly six faces to avoid errors and ensure accurate processing.
- Use the padding_mode parameter to adjust the interpolation method if you notice artifacts or quality issues in the output image.
- Experiment with different cube_format settings to see which arrangement best suits your specific image processing needs.
Masked Diff C2E Common Errors and Solutions:
"Only a batch size of 1 is currently supported for original_equi"
- Explanation: This error occurs when the original_equi parameter has a batch size greater than 1. - Solution: Ensure that the original_equi tensor is provided with a batch size of exactly 1.
"original_faces should have 6 faces"
- Explanation: This error indicates that the original_faces tensor does not contain the required six faces.
- Solution: Verify that the original_faces input consists of exactly six faces, corresponding to the six sides of a cube.
"modified_faces should have 6 faces"
- Explanation: This error suggests that the modified_faces tensor does not have the necessary six faces.
- Solution: Check that the modified_faces input includes exactly six faces to match the cube map format.