A dependency-free (pure standard-library) Python companion that computes NO metric itself. --registry prints the open-source no-reference toolbox catalogue (FFmpeg built-in NR filters, CAMBI, pyiqa/IQA-PyTorch, scikit-video, FAST-VQA, DOVER, Google UVQ, itu-p1203) with what each measures, install command, reference (always none), licence, and reported accuracy. --recommend takes needs as flags (--use {live,ugc,vod-screening,banding,blockiness,blur,per-frame,session}, --gpu, --commercial, --laten
Download free PDF
The handbook · 7 volumes
Fora Soft Video Engineering Handbook
This no-reference toolbox advisor catalogues open-source blind metrics from FFmpeg filters and CAMBI to DOVER and UVQ, and recommends one for your use case, hardware and licence. Our seven-volume handbook covers the video stack; start with Volume 6 on AI models, inference and deployment topologies, then Volume 1 on quality metrics and Volume 3 on QoE monitoring. Written by Nikolay Sapunov, CEO at Fora Soft.
Picking an open-source no-reference metric for production monitoring?
Describe your content, latency and GPU budget, and we will shortlist no-reference tools and plan how to validate them against viewer ratings. 250+ video products shipped since 2005.
You get a realistic estimate, technical recommendations and next steps. Free, no obligations.
