ComfyUI Node: PanoLongMaskSplit

PanoLongMaskSplit:

The PanoLongMaskSplit node is designed to handle panoramic masks by splitting a long mask into individual face masks, which are essential for processing panoramic images. This node is particularly useful in scenarios where you need to manipulate or analyze different sections of a panoramic image separately. By splitting the long mask, it allows for more granular control over each face of the panorama, enabling you to apply different effects or transformations to each section. This capability is crucial for artists and developers working with 360-degree images or virtual reality content, as it provides the flexibility to manage and edit each face of the panorama independently.

PanoLongMaskSplit Input Parameters:

long_mask

The long_mask parameter is a crucial input that represents the panoramic mask to be split. It is expected to be a tensor with a specific shape where the width is six times the height, corresponding to the six faces of a cube map. This parameter serves as the foundation for the node's operation, as it determines the initial mask that will be divided into individual face masks.

ratio

The ratio parameter is a floating-point value that dictates the size of the mask to be extracted from each face of the panorama. It ranges from 0.1 to 1.0, with a default value of 1.0. This parameter allows you to control the proportion of each face that will be masked, providing flexibility in how much of each face is included in the final output. A smaller ratio results in a smaller mask, while a ratio of 1.0 uses the full face.

invert

The invert parameter is a boolean option that determines whether the mask should be inverted. By default, it is set to false, meaning the mask is used as-is. When set to true, the mask is inverted, which can be useful for operations where you need to exclude certain areas rather than include them. This parameter provides additional control over the mask's application, allowing for more complex masking operations.

PanoLongMaskSplit Output Parameters:

mask

The mask output is a single tensor that represents the combined mask for the entire panorama. It is constructed by applying the specified ratio and inversion settings to each face of the long mask. This output is useful for scenarios where you need a unified mask that covers the entire panoramic image, allowing for consistent processing across all faces.

masks

The masks output is a list of tensors, each corresponding to a face of the panorama. This output provides the individual face masks, which are essential for applying different effects or transformations to each section of the panorama. By having access to these separate masks, you can achieve more detailed and customized editing of panoramic images.

PanoLongMaskSplit Usage Tips:

• Ensure that the long_mask input has the correct shape, with the width being six times the height, to avoid errors and ensure proper mask splitting.
• Use the ratio parameter to control the size of the mask applied to each face. A smaller ratio can be useful for focusing on specific areas, while a ratio of 1.0 covers the entire face.
• Consider using the invert parameter when you need to exclude certain areas from the mask, providing more flexibility in your masking operations.

PanoLongMaskSplit Common Errors and Solutions:

Mask shape error.

• Explanation: This error occurs when the long_mask input does not have the expected shape, specifically when the width is not six times the height.
• Solution: Verify that the long_mask input is correctly formatted, with the width being exactly six times the height, to ensure proper mask splitting.

The number of masks must be 6.

• Explanation: This error indicates that the input or output masks do not contain the expected six faces, which is necessary for processing a full panorama.
• Solution: Check that the input mask is correctly structured and that the node's operations are producing the expected six face masks. Adjust the input or node settings as needed to achieve the correct output.