Skip to content

JioJe/comfyui_video_BC

Repository files navigation

comfyui_video_BC

这是一个针对ComfyUI的视频工作流程的自定义节点包,包含两个节点:一个用于批量加载视频帧,另一个用于将图像序列保存为视频,具有自定义设置。

🧩 包含的节点

##新增节点(2025/7/14)

  1. 批量文本读取(TextBatchLoader)

功能:从指定文件夹一次性读取所有 .txt 文件,输出文本内容列表和对应文件名列表。适用于批量文本数据的自动化处理场景。

  1. 批量文本索引(TextBatchIndexer)

功能:根据输入的索引,从批量文本和文件名列表中提取指定位置的文本内容和文件名。适合逐条遍历、单条处理等需求。

  1. 批量文本保存(TextBatchSaver)

功能:将批量文本内容保存到指定目录,支持自定义文件名前缀、分隔符、数字填充、扩展名、编码、后缀等。可选择保留原文件名或自定义命名,输出所有保存文件的路径列表。

  1. 批量提示词替换(TextBatchReplace)

功能:对批量文本内容进行多组关键词批量替换。支持在界面中直观填写多组“查找/替换”对,实现高效的批量文本内容修改。

1. VideoSequenceProcessor

  • 新增设置导入视频强制帧率选项
  • 从目录中的多个视频加载帧。
  • 模式:按索引加载、自动下一个或随机选择。
  • 可选的尺寸调整、帧限制和帧率覆盖。
  • 适用于数据集准备和动画工作流程。

2. VideoCombine

  • 将图像序列(例如来自AnimateDiff的图像)转换为视频。
  • 支持多种格式:MP4、AVI、MKV、MOV、WMV。
  • 可选的“乒乓”循环。
  • 可自定义文件名和输出路径。

📦 安装

将这些Python文件放入ComfyUI/custom_nodes目录中。

🔧 依赖

确保已安装OpenCV和其他依赖项:

git clone https://github.com/JioJe/comfyui_video_BC

B站节点讲解链接:https://www.bilibili.com/video/BV1heZVY3E8o/?spm_id_from=333.1387.homepage.video_card.click&vd_source=cfcf2acc406e911d0270e4c5832b80a8

These are two custom nodes for ComfyUI designed to streamline video workflows: one for batch loading video frames and one for saving image sequences as videos with custom settings.

🧩 Nodes Included

NEW

  1. TextBatchLoader(2025/7/14)

Function: Loads all .txt files from a specified folder in batch, outputs a list of text contents and corresponding filenames. Ideal for automated batch text processing.

  1. TextBatchIndexer

Function: Retrieves the text and filename at a specified index from batch lists. Useful for iterating or processing items one by one.

  1. TextBatchSaver

Function: Saves a batch of text contents to a target directory, supporting custom filename prefix, separator, zero-padding, extension, encoding, and suffix. Can use original filenames or custom naming. Outputs the list of saved file paths.

  1. TextBatchReplace

Function: Performs multiple keyword replacements on a batch of texts. Supports intuitive UI for entering multiple search/replace pairs, enabling efficient batch text modification.

1. VideoSequenceProcessor

  • Load frames from multiple videos in a directory.
  • Modes: load by index, auto-next, or random selection.
  • Optional resize, frame limit, and frame rate override.
  • Ideal for dataset preparation and animation workflows.

2. VideoCombine

  • Convert image sequences (e.g., from AnimateDiff) into a video.
  • Supports multiple formats: MP4, AVI, MKV, MOV, WMV.
  • Optional "ping-pong" looping.
  • Customizable filename and output path.

📦 Installation

Place these Python files into your ComfyUI/custom_nodes directory.

🔧 Requirements

Ensure you have OpenCV and other dependencies installed:

git clone https://github.com/JioJe/comfyui_video_BC

About

ComfyUI Batch Video Loader & Video Combine Nodes

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages