The Intel® Media SDK

桌面共享, 支持RTSP转发, RTSP推流, RTMP推流。

Shadowplay's Replay Feature On Linux For Nvidia, AMD and Intel

Set of flows for Tdarr. Also lots of ffmpeg resources on how to use various encoders

This collection of samples demonstrates best practices to achieve optimal video quality and performance on Intel GPUs for content delivery networks. Check out our demo, recommended command lines and quality and performance measuring tools.

A tool developed in Rust to convert QSV to FLV（一个爱奇艺qsv转flv工具，不会生成多余的临时文件，使用Rust开发）

hardware codec for windows using ffmpeg (nvi qsv enc\dxva2 dec) 使用ffmpeg进行Qsv硬编码、Nvi硬编码、DXVA硬解码的封装API

Client for the Jellyfin Hardware Survey Server (https://hwa.jellyfin.org/). Benchmarks hardware performance for simultaneous ffmpeg transcoding, enabling detailed comparisons and insights for optimizing Jellyfin setups.

Fast cross-compile ffmpeg for Windows with MinGW on Linux

Rust bindings to Intel MediaSDK

CSV sniffer crate for Rust, optimized for qsv

ミニPC向けのよくばりセット / Mirakurun + recpt1 + EPGStation + ffmpeg + qsv on Docker

FFmpeg build script using Conan 1.X, targeting hardware acceleration. Without changes, the resulting binary matches the one used in the latest TeamSpeak 6 Client.

An attempt at an automated tool for finding the best possible media quality an encoder and codec can give.

🌐 Promo website for qsv pro, a spreadsheet data wrangling desktop app. Includes download links for Windows, macOS, & Linux. Website built with Astro as a static site.

使用ffmpeg将h264视频转换为h265

A dockerized FFmpeg compiled with non-free codecs in.

Compresses your videos down to a set size

VMAF를 사용하여 비디오 품질과 파일 크기 사이의 최적점(sweet spot)을 찾는 자동 벤치마킹 GUI 애플리케이션. (소프트웨어 및 하드웨어(NVENC, QSV, AMF) 인코더 지원) / An automated benchmarking GUI application that uses VMAF to find the sweet spot between video quality and file size. (Supports software and hardware (NVENC, QSV, AMF) encoders.)

🧩 (WIP) The book "100 exercises with qsv". For new qsv users to read lessons and try out exercises either in-browser or locally in their terminal. Built with Jupyter Book.