Visit ComfyUI Online for ready-to-use ComfyUI environment
Address flickering in image sequences with temporal smoothing and adaptive blending for stable, high-quality visuals.
Pixel Deflicker (SuperBeasts.AI) is designed to address the common issue of flickering in sequences of images or video frames. Flickering can occur due to variations in lighting, exposure, or other inconsistencies between frames, which can be distracting and reduce the overall quality of the visual content. This node applies temporal smoothing and adaptive blending techniques to minimize these inconsistencies, resulting in a more stable and visually appealing output. By leveraging advanced image processing methods, Pixel Deflicker ensures that the transitions between frames are smooth, enhancing the viewing experience without compromising the original quality of the images.
This parameter takes a sequence of images that you want to process to remove flickering. The images should be provided as a list or batch of tensors. The quality and consistency of these input images directly impact the effectiveness of the deflickering process.
The window size parameter determines the number of frames considered for temporal smoothing. A larger window size can result in smoother transitions but may also introduce more latency in the changes. The default value is 5, which provides a balance between smoothness and responsiveness. Minimum value is 1, and there is no strict maximum, but very large values may affect performance.
Blending strength controls the degree to which the smoothed frames are blended with the original frames. A value of 0 means no blending (original frames are retained), while a value of 1 means full blending (only smoothed frames are used). The default value is 0.5, which provides a balanced blend. The range is from 0 to 1.
Batch size specifies the number of frames processed together in one batch. Processing in batches can improve performance, especially for large sequences of images. The default value is 10, which is a good starting point for most use cases. Adjusting this value can help optimize performance based on your hardware capabilities.
The output is a tensor containing the deflickered sequence of images. This tensor represents the processed frames where flickering has been minimized, resulting in a smoother visual experience. The output retains the original dimensions and quality of the input images, ensuring that the deflickering process does not degrade the visual content.
ValueError: Input images must be a list of tensors
IndexError: List index out of range
TypeError: Unsupported image format
© Copyright 2024 RunComfy. All Rights Reserved.