Dual Branch Video Quality Evaluation Method

Human beings predict video quality scores by simulating the perception mechanism of the Human Visual System (HVS), which has been proven to be an effective way to solve the problem of Video Quality Assessment (VQA). This article presents a new dual branch video quality evaluation method. Firstly, sample the video frames using the designed sampling method. Next, two parallel independent feature extractors are used for deep extraction of video features. These two extractors focus on different aspects of the video without interfering with each other, thus capturing the intrinsic features of the video more targeted. Then, feature fusion was achieved through the designed attention mechanism, further enhancing the model's expressive power. Finally, the features are mapped into accurate and reliable video quality scores through methods such as spatial attention and temporal attention. This article provides a comprehensive and efficient video quality assessment framework, proposing new ideas and methods for VQA.