SimpleVQA and CompressedVQA
About SimpleVQA
sunwei925/SimpleVQA
A Deep Learning based No-reference Quality Assessment Model for UGC Videos
This tool helps content creators and platforms automatically assess the quality of user-generated videos without needing a "perfect" reference version. You feed it a user-uploaded video, and it outputs a quality score indicating how good or bad the video looks. It's designed for anyone managing or analyzing large volumes of user-submitted video content.
About CompressedVQA
sunwei925/CompressedVQA
Deep Learning based Full-reference and No-reference Quality Assessment Models for Compressed UGC Videos
This project helps video platform managers and content creators evaluate the visual quality of compressed user-generated content (UGC) videos. By inputting either a compressed video alongside its original (for full-reference) or just the compressed video (for no-reference), it outputs a quality score. This is useful for anyone responsible for video quality on platforms like YouTube, TikTok, or streaming services.
Scores updated daily from GitHub, PyPI, and npm data. How scores work