Adaptive Motion Compensated Temporal Filter for Video Noise Reduction

The patent applies dynamics by making the filter strength adaptive rather than fixed. The filter strength is dynamically adjusted based on the motion vector magnitude and temporal correlation characteristics of the video signal. When motion is detected, the filter strength is reduced to preserve signal content, while when motion is minimal, the filter strength is increased to maximize noise suppression. This dynamic adaptation allows the system to remove more noise effectively without wasting bandwidth on already-clean signals.

The patent changes the parameter of filter strength based on detected motion characteristics. By monitoring motion vectors and temporal correlation, the system adjusts the filtering parameters in real-time. The filter transitions between different strength levels (e.g., weak filtering when motion is present, strong filtering when motion is minimal), enabling optimal noise removal while preserving signal integrity and avoiding unnecessary bandwidth consumption on clean frames.

The patent applies local quality by differentiating between different regions of the video signal based on motion characteristics. Areas with high motion vectors receive weaker filtering to preserve signal content, while areas with low motion vectors receive stronger filtering to remove noise. This spatially and temporally adaptive approach ensures that noise suppression is applied locally where appropriate without disrupting signal content in motion-prone regions.

The system dynamically adjusts filter strength based on real-time motion detection. When motion is detected in a particular region or frame, the filter strength is automatically reduced to preserve signal content. When motion is minimal, the filter strength increases to maximize noise suppression. This dynamic adaptation resolves the contradiction by making the filtering behavior context-dependent rather than static.