HUB
Sign inSign up

hexstan/video-watchdog

By hexstan

•Updated 8 months ago

Image
0

1.1K

hexstan/video-watchdog repository overview

⁠VideoWatchdog

GitHub Repository⁠ | Issue Tracker⁠

VideoWatchdog is a lightweight Python-based automation tool designed to monitor specified directories for video files and automatically invoke FFmpeg for processing once file writing is complete. It is ideal for scenarios requiring automated video transcoding, compression, or format conversion (e.g., integrated with NAS, downloaders, or other automated workflows).

⁠✨ Key Features

  • Directory Monitoring: Supports continuous directory monitoring or running as a one-time script.
  • Multi-task Support: Define multiple independent monitoring and processing tasks within a single configuration file.
  • File Stability Detection: Ensures only fully written files are processed by tracking modification times and file size changes, preventing the handling of incomplete downloads or copies.
  • Custom FFmpeg Commands: Supports fully customizable FFmpeg processing commands to flexibly meet various audio/video processing needs.
  • Hardware Acceleration: Supports GPU-accelerated transcoding via Intel QSV, AMD VAAPI, and NVIDIA NVENC.
  • Auto-Archiving & Status Management: Automatically archives source files after processing and records status to prevent duplicate processing.

⁠🚀 How to Use This Image

See GitHub Repo⁠

⁠⚙️ Hardware Acceleration Configuration (GPU)

If you need to use a GPU for hardware-accelerated video transcoding, please map the corresponding hardware devices upon startup.

⁠Intel / AMD GPU (VAAPI / QSV)

Add the devices mapping in your docker-compose.yml:

    devices:
      - /dev/dri:/dev/dri

Or add --device /dev/dri:/dev/dri to your docker run command.

⁠NVIDIA GPU (NVENC)

Ensure the NVIDIA Container Toolkit⁠ is installed on the host machine.

Add the deploy configuration in your docker-compose.yml:

    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu, video, compute, utility]

Or add --gpus all to your docker run command.


⁠VideoWatchdog

GitHub Repository⁠ | Issue Tracker⁠

VideoWatchdog 是一个基于 Python 的轻量级自动化工具,用于监控指定目录中的视频文件,并在文件写入完成后自动调用 FFmpeg 进行处理。它非常适合需要自动化视频转码、压缩或格式转换的场景(例如配合 NAS、下载器等自动化工作流使用)。

⁠✨ 核心特性

  • 目录监控:支持持续监控目录或作为一次性脚本运行。
  • 多任务支持:可以在单个配置文件中定义多个独立的监控和处理任务。
  • 文件稳定性检测:通过检测文件修改时间和大小变化,确保只处理已完全写入的文件,避免处理正在下载或复制中的不完整文件。
  • 自定义 FFmpeg 命令:支持完全自定义的 FFmpeg 处理命令,灵活应对各种音视频处理需求。
  • 硬件加速支持:支持 Intel QSV、AMD VAAPI 以及 NVIDIA NVENC 等 GPU 硬件加速转码。
  • 自动归档与状态管理:处理完成后自动归档源文件,并记录状态防止重复处理。

⁠🚀 如何使用此镜像

见GitHub仓库⁠

⁠⚙️ 硬件加速配置 (GPU)

如果你需要使用 GPU 进行视频转码加速,请在启动时映射相应的硬件设备。

⁠Intel / AMD GPU (VAAPI / QSV)

在 docker-compose.yml 中添加 devices 映射:

    devices:
      - /dev/dri:/dev/dri

或者在 docker run 命令中添加 --device /dev/dri:/dev/dri。

⁠NVIDIA GPU (NVENC)

确保宿主机已安装 NVIDIA Container Toolkit⁠。

在 docker-compose.yml 中添加 deploy 配置:

    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu, video, compute, utility]

或者在 docker run 命令中添加 --gpus all。

Tag summary

Content type

Image

Digest

sha256:caf120c66…

Size

231.8 MB

Last updated

8 months ago

docker pull hexstan/video-watchdog