VQTDL (Video Quality Testing with Deep Learning)

Video Quality Testing with Deep Learning—or VQTDL—is a no-reference algorithm for video quality assessment. This solution produces image quality predictions that correlate well with human perception and offers good performance under diverse circumstances, such as various network conditions, platforms and applications.

Example: A company is updating the animated login screen on its website. The quality assurance team wants to ensure the video loads quickly without sacrificing visual quality. They use a VQTDL-based tool to analyze the compressed video. The deep learning model evaluates the video stream to automatically detect any noticeable artifacts, dropped frames, or color distortions introduced by the compression. It then provides a clear score, such as a Video Mean Opinion Score (V-MOS), which tells the team if the video quality is high enough to be acceptable to the user. This objective data helps them find the optimal balance between a small file size and a great user experience.