Machine learning based chroma phase offset detection and correction in motion video

<span lang="EN-US">Generally, chroma phase or hue offset issues within a scene are hard to detect, without reference context (i.e. some apriori<em> </em>knowledge about how certain objects the should actually<em> </em>appear in terms of their hue). Moreover, when it comes skin/flesh tones, deviation can be noticeable and markedly degrade viewer quality experience</span><span lang="IN">(QoE)</span><span lang="EN-US">, whenever does occur. However lot research has gone into flesh tone detection, specifically, color gamut which is present. This topic been well documented literature with respect various spaces: red, green, blue (RGB) YIQ. Therefore, overall video content could potentially approached by extracting analyzing reliable reference, such as skin (if present), allowable deviation. involves machine learning (ML) based facial recognition tracking followed region detected sequence Region Interest). The serves ‘self-reference’ order discern any inherent content. Finally, angular discerned then used for subsequent correction well.</span><p> </p>