init

.gitignore

@@ -0,0 +1,2 @@
+/fonts
+/__pycache__

README.md

@@ -0,0 +1,91 @@
+<h1 align="center">ComfyUI-FFmpeg</h1>
+
+<p align="center">
+    <br> <font size=5>中文 | <a href="README_EN.md">English</a></font>
+</p>
+
+
+## 介绍
+
+把FFmpeg常用功能封装成ComfyUI节点，方便用户可以在ComfyUI上也可以进行各种视频处理。<br>
+
+## 安装 
+
+#### 方法1:
+
+1. 进入节点目录, `ComfyUI/custom_nodes/`
+2. `git clone https://github.com/MoonHugo/ComfyUI-FFmpeg.git`
+3. `cd ComfyUI-FFmpeg`
+4. `pip install -r requirements.txt`
+5. 重启ComfyUI
+
+#### 方法2:
+直接下载节点源码包，然后解压到custom_nodes目录下，最后重启ComfyUI
+
+#### 方法3：
+通过ComfyUI-Manager安装，搜索“ComfyUI-FFmpeg”进行安装
+
+## 节点介绍
+
+##### Video2Frames节点: 作用是将视频转为一张一张的图片，并保存到指定目录中<br>
+
+![](./assets/1.png)
+
+###### 参数说明
+**video_path**: 本地视频路径，比如：`C:\Users\Desktop\222.mp4`
+**output_path**: 输出图片保存路径，比如：`C:\Users\Desktop\output`
+**frames_max_width**: 这个参数可以用来缩放视频，默认为0，表示不缩放视频，如果frames_max_width大于视频实际宽度，则视频不会被放大，保持原宽度，如果frames_max_width小于视频实际宽度，则视频会被缩小。
+
+___
+
+##### Frames2Video节点: 作用是将图片转为视频，并保存到指定目录中<br>
+![](./assets/2.png)
+
+###### 参数说明
+**frame_path**: 本地图片路径，比如：`C:\Users\Desktop\output`
+**fps**: 视频帧率，默认为`30`
+**video_name**: 保存视频名称，比如：`222.mp4`
+**output_path**: 视频保存路径，比如：`C:\Users\Desktop\output`
+**audio_path**: 视频音频路径，比如：`C:\Users\Desktop\222.mp3`
+___
+
+##### AddTextWatermark节点: 作用是在视频上添加文字水印<br>
+
+![](./assets/3.png)
+
+###### 参数说明
+**video_path**: 本地视频路径，比如：`C:\Users\Desktop\222.mp4`
+**output_path**: 视频保存路径，比如：`C:\Users\Desktop\output`
+**font_file**: 字体文件，需要把字体文件放到`custom_nodes\ComfyUI-FFmpeg\fonts`目录下，不仅英文字体，中文字体也可以，比如：`ComfyUI\custom_nodes\ComfyUI-FFmpeg\fonts\Alibaba-PuHuiTi-Heavy.ttf`
+**font_size**: 水印文字大小，比如：`40`
+**font_color**: 水印文字颜色，比如：`#FFFFFF`或者`white`
+**position_x**: 水印文字x坐标，比如：`100`
+**position_y**: 水印文字y坐标，比如：`100`
+
+___
+
+##### AddImgWatermark节点: 作用是在视频上添加图片水印<br>
+
+![](./assets/4.png)
+
+###### 参数说明
+**video_path**: 本地视频路径，比如：`C:\Users\Desktop\222.mp4`
+**output_path**: 视频保存路径，比如：`C:\Users\Desktop\output`
+**watermark_image**: 水印图片路径，比如：`C:\Users\Desktop\watermark.png`
+**watermark_img_width**: 水印图片宽度，比如：`100`
+**position_x**: 水印图片在视频中的x坐标，比如：`100`
+**position_y**: 水印图片在视频中的y坐标，比如：`100`
+___
+
+## 社交账号
+- Bilibili：[我的B站主页](https://space.bilibili.com/1303099255)
+
+## 感谢
+
+感谢FFmpeg仓库的所有作者 [FFmpeg/FFmpeg](https://github.com/FFmpeg/FFmpeg)
+
+部分代码参考了 [Eden-yidun/ComfyUI-EdenVideoTool](https://github.com/Eden-yidun/ComfyUI-EdenVideoTool) 感谢！
+
+## 关注历史
+
+[![Star History Chart](https://api.star-history.com/svg?repos=MoonHugo/ComfyUI-FFmpeg&type=Date)](https://star-history.com/#MoonHugo/ComfyUI-FFmpeg&Date)

README_EN.md

@@ -0,0 +1,91 @@
+<h1 align="center">ComfyUI-FFmpeg</h1>
+
+<p align="center">
+    <br> <font size=5>English | <a href="README.md">中文</a></font>
+</p>
+
+
+## Introduction
+
+Encapsulate the commonly used functions of FFmpeg into ComfyUI nodes, making it convenient for users to perform various video processing tasks within ComfyUI.<br>
+
+## Installation 
+
+#### Method 1:
+
+1. Go to comfyUI custom_nodes folder, `ComfyUI/custom_nodes/`
+2. `git clone https://github.com/MoonHugo/ComfyUI-FFmpeg.git`
+3. `cd ComfyUI-FFmpeg`
+4. `pip install -r requirements.txt`
+5. restart ComfyUI
+
+#### Method 2:
+Directly download the node source package, then extract it into the custom_nodes directory, and finally restart ComfyUI.
+
+#### Method 3：
+Install through ComfyUI-Manager by searching for 'ComfyUI-BiRefNet-Hugo' and installing it.
+
+## Nodes introduction
+
+##### Video2Frames Node: The function is to convert a video into images and save them to a specified directory.<br>
+
+![](./assets/1.png)
+
+###### Parameter Description
+**video_path**: The local video path, e.g：`C:\Users\Desktop\222.mp4`
+**output_path**: The path to save the output images, e.g：`C:\Users\Desktop\output`
+**frames_max_width**: This parameter can be used to resize the video. The default value is 0, which means the video will not be resized. If frames_max_width is larger than the actual width of the video, the video will not be enlarged and will retain its original width. If frames_max_width is smaller than the actual width of the video, the video will be scaled down.
+
+___
+
+##### Frames2Video Node: The function is to convert images into a video and save it to a specified directory.<br>
+![](./assets/2.png)
+
+###### Parameter Description
+**frame_path**: local image path, e.g:`C:\Users\Desktop\output`
+**fps**: video frame rate, default is`30`
+**video_name**: saved video name, e.g:`222.mp4`
+**output_path**: video save path,e.g:`C:\Users\Desktop\output`
+**audio_path**: video audio path,e.g:`C:\Users\Desktop\222.mp3`
+___
+
+##### AddTextWatermark Node: The function is to add a text watermark to the video.<br>
+
+![](./assets/3.png)
+
+###### Parameter Description
+**video_path**: local video path,e.g:`C:\Users\Desktop\222.mp4`
+**output_path**: video save path,e.g:`C:\Users\Desktop\output`
+**font_file**: font file: The font file needs to be placed in the`custom_nodes\ComfyUI-FFmpeg\fonts` directory. Not only English fonts, but Chinese fonts can also be used.,e.g:`ComfyUI\custom_nodes\ComfyUI-FFmpeg\fonts\Alibaba-PuHuiTi-Heavy.ttf`
+**font_size**: watermark text size,e.g:`40`
+**font_color**: watermark text color,e.g:`#FFFFFF` or `white`
+**position_x**: watermark text x-coordinate,e.g:`100`
+**position_y**: watermark text y-coordinate,e.g:`100`
+
+___
+
+##### AddImgWatermark Node: The function is to add an image watermark to the video.<br>
+
+![](./assets/4.png)
+
+###### Parameter Description
+**video_path**: local video path,e.g:`C:\Users\Desktop\222.mp4`
+**output_path**: video save path,e.g:`C:\Users\Desktop\output`
+**watermark_image**: watermark image path,e.g:`C:\Users\Desktop\watermark.png`
+**watermark_img_width**: watermark image width,e.g:`100`
+**position_x**: watermark image x-coordinate in the video,e.g:`100`
+**position_y**: watermark image y-coordinate in the video,e.g:`100`
+___
+
+## Social Account Homepage
+- Bilibili：[My BILIBILI Homepage](https://space.bilibili.com/1303099255)
+
+## Acknowledgments
+
+Thanks to all the contributors of the FFmpeg repository. [FFmpeg/FFmpeg](https://github.com/FFmpeg/FFmpeg)
+
+Some of the code references [Eden-yidun/ComfyUI-EdenVideoTool](https://github.com/Eden-yidun/ComfyUI-EdenVideoTool) Thanks!
+
+## Star history
+
+[![Star History Chart](https://api.star-history.com/svg?repos=MoonHugo/ComfyUI-FFmpeg&type=Date)](https://star-history.com/#MoonHugo/ComfyUI-FFmpeg&Date)

__init__.py

@@ -0,0 +1,3 @@
+from .nodes_map import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
+
+__all__ = ["NODE_CLASS_MAPPINGS", "NODE_DISPLAY_NAME_MAPPINGS"]

func.py

@@ -0,0 +1,48 @@
+import numpy as np
+from PIL import Image
+import torch
+import subprocess
+import json
+
+def tensor2pil(image):
+    return Image.fromarray(np.clip(255. * image.cpu().numpy().squeeze(), 0, 255).astype(np.uint8))
+
+def pil2tensor(image):
+    return torch.from_numpy(np.array(image).astype(np.float32) / 255.0).unsqueeze(0)
+
+def getVideoInfo(video_path):
+    command = [
+            'ffprobe', '-v', 'error', '-select_streams', 'v:0', '-show_entries', 
+            'stream=avg_frame_rate,duration,width,height', '-of', 'json', video_path
+        ]
+    # 运行ffprobe命令
+    result = subprocess.run(command, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
+    # 将输出转化为字符串
+    output = result.stdout.decode('utf-8').strip()
+    print(output)
+    data = json.loads(output)
+    # 查找视频流信息
+    if 'streams' in data and len(data['streams']) > 0:
+        stream = data['streams'][0]  # 获取第一个视频流
+        fps = stream.get('avg_frame_rate')
+        if fps is not None:
+            # 帧率可能是一个分数形式的字符串，例如 "30/1" 或 "20.233000"
+            if '/' in fps:
+                num, denom = map(int, fps.split('/'))
+                fps = num / denom
+            else:
+                fps = float(fps)  # 直接转换为浮点数
+            width = int(stream.get('width'))
+            height = int(stream.get('height'))
+            duration = float(stream.get('duration'))
+            return_data = {'fps': fps, 'width': width, 'height': height, 'duration': duration}
+    else:
+        return_data = {}
+    return return_data
+
+def get_image_size(image_path):
+    # 打开图像文件
+    with Image.open(image_path) as img:
+        # 获取图像的宽度和高度
+        width, height = img.size
+        return width, height

nodes/addImgWatermark.py

@@ -0,0 +1,79 @@
+import os
+from PIL import Image
+import subprocess
+import time
+from ..func import get_image_size
+
+
+class AddImgWatermark:
+ 
+    # 初始化方法
+    def __init__(self): 
+        pass 
+    
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+            "required": { 
+                "video_path": ("STRING", {"default":"C:/Users/Desktop/video.mp4",}),
+                "output_path": ("STRING", {"default":"C:/Users/Desktop/output/",}),
+                "watermark_image": ("STRING", {"default":"C:/Users/Desktop/logo.png",}),
+                "watermark_img_width":  ("INT", {"default": 100,"min": 1, "step": 1}),
+                "position_x":  ("INT", {"default": 10, "step": 1}),
+                "position_y":  ("INT", {"default": 10, "step": 1}),
+            },
+        }
+
+    RETURN_TYPES = ("STRING",)
+    RETURN_NAMES = ("video_complete_path",)
+    FUNCTION = "add_img_watermark" 
+    OUTPUT_NODE = True
+    CATEGORY = "FFmpeg" 
+
+    def add_img_watermark(self,video_path,output_path,watermark_image,watermark_img_width,position_x,position_y):
+        try:
+            # 视频不存在
+            if not video_path.lower().endswith(('.mp4', '.avi', '.mov', '.mkv','.rmvb')):
+                raise ValueError("video_path不是视频文件（video_path is not a video file）")
+            
+            if not os.path.exists(video_path):
+                raise ValueError("video_path不存在（video_path does not exist）")
+            
+            #判断output_path是否是一个目录
+            if not os.path.isdir(output_path):
+                raise ValueError("output_path不是目录（output_path is not a directory）")
+            
+            # 文件不是图片
+            if not watermark_image.lower().endswith(('.png', '.jpg', '.jpeg', '.bmp', '.gif')):
+                raise ValueError("水印文件不是图片文件（watermark file is not a image file）")
+            
+            if not os.path.exists(watermark_image):
+                raise ValueError("水印文件不存在（watermark file does not exist）")
+            
+            file_name = os.path.basename(video_path)
+            file_extension = os.path.splitext(file_name)[1]
+            #文件名根据年月日时分秒来命名
+            file_name = time.strftime("%Y%m%d%H%M%S", time.localtime()) + file_extension
+            output_path = os.path.join(output_path, file_name)
+            width,height = get_image_size(watermark_image)
+            watermark_img_height = int(height * watermark_img_width / width)  # 按比例计算新高度
+            cmd = [
+                'ffmpeg',
+                '-i', video_path,
+                '-i', watermark_image,
+                '-filter_complex',f"[1:v]scale={watermark_img_width}:{watermark_img_height}[wm];[0:v][wm]overlay=x={position_x}:y={position_y}:format=auto",
+                output_path,
+            ]
+            result = subprocess.run(cmd, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
+            #os.remove(image_save_path) # 删除临时水印图片
+            # 检查返回码
+            if result.returncode != 0:
+                # 如果有错误，输出错误信息
+                 print(f"Error: {result.stderr.decode('utf-8')}")
+                 raise ValueError(f"Error: {result.stderr.decode('utf-8')}")
+            else:
+                # 输出标准输出信息
+                print(result.stdout)
+        except Exception as e:
+            raise ValueError(e)
+        return (output_path,)

nodes/addTextWatermark.py

@@ -0,0 +1,85 @@
+import os
+import subprocess
+import folder_paths
+import time
+
+current_path = os.path.abspath(__file__)
+font_dir = os.path.join(os.path.dirname(os.path.dirname(os.path.normpath(__file__))), 'fonts')
+folder_paths.folder_names_and_paths["fonts"] = ([font_dir], {'.ttf'})
+
+class AddTextWatermark:
+ 
+    # 初始化方法
+    def __init__(self): 
+        pass 
+    
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+            "required": { 
+                "video_path": ("STRING", {"default":"C:/Users/Desktop/video.mp4",}),
+                "output_path": ("STRING", {"default":"C:/Users/Desktop/output/",}),
+                'font_file': (["default"] + folder_paths.get_filename_list("fonts"), ),
+                'font_size': ("INT", {"default": 15, "min": 1, "max": 1000, "step": 1}),
+                'font_color': ("STRING", {"default": "#FFFFFF"}),
+                "text": ("STRING", {"default": "Watermark"}),
+                "position_x":  ("INT", {"default": 10, "step": 1}),
+                "position_y":  ("INT", {"default": 10, "step": 1}),
+            },
+        }
+
+    RETURN_TYPES = ("STRING","STRING",)
+    RETURN_NAMES = ("video_path","output_path",)
+    FUNCTION = "add_text_watermark" 
+    OUTPUT_NODE = True
+    CATEGORY = "FFmpeg" 
+
+    def add_text_watermark(self,video_path,output_path,font_file,font_size,font_color,text,position_x,position_y):
+        try:
+             # 视频不存在
+            if not video_path.lower().endswith(('.mp4', '.avi', '.mov', '.mkv','.rmvb')):
+                raise ValueError("video_path不是视频文件（video_path is not a video file）")
+            
+            if not os.path.exists(video_path):
+                raise ValueError("video_path不存在（video_path does not exist）")
+            
+            #判断output_path是否是一个目录
+            if not os.path.isdir(output_path):
+                raise ValueError("output_path不是目录（output_path is not a directory）")
+            
+            file_name = os.path.basename(video_path)
+            file_extension = os.path.splitext(file_name)[1]
+            #文件名根据年月日时分秒来命名
+            file_name = time.strftime("%Y%m%d%H%M%S", time.localtime()) + file_extension
+            output_path = os.path.join(output_path, file_name)
+            
+            # 替换为双斜杠
+            font_path = os.path.join(font_dir, font_file).replace("\\", "/").replace(":", "\\:")
+            # 构建命令 C\\:/Windows/Fonts/simhei.ttf   fontfile='J\\:/Comfyui-for-OOTDiffusion/ComfyUI/custom_nodes/ComfyUI-FFmpeg/fonts/Alibaba-PuHuiTi-Heavy.ttf
+            if font_file == "default":
+                cmd = [
+                    'ffmpeg',
+                    '-i', video_path,
+                    '-vf', f"drawtext=text='{text}':x={position_x}:y={position_y}:fontsize={font_size}:fontcolor={font_color}",
+                    output_path,
+                ]
+            else:
+                cmd = [
+                    'ffmpeg',
+                    '-i', video_path,
+                    '-vf', f"drawtext=text='{text}':x={position_x}:y={position_y}:fontfile='{font_path}':fontsize={font_size}:fontcolor={font_color}",
+                    output_path,
+                ]
+            result = subprocess.run(cmd, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
+            # 检查返回码
+            if result.returncode != 0:
+                # 如果有错误，输出错误信息
+                 print(f"Error: {result.stderr.decode('utf-8')}")
+                 raise ValueError(f"Error: {result.stderr.decode('utf-8')}")
+            else:
+                # 输出标准输出信息
+                print(result.stdout)
+        except Exception as e:
+            raise ValueError(e)
+        
+        return (video_path,output_path)

nodes/frames2video.py

@@ -0,0 +1,125 @@
+import os
+import subprocess
+import re
+from PIL import Image
+from pathlib import Path
+
+class Frames2Video:
+ 
+    # 初始化方法
+    def __init__(self): 
+        pass 
+    
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+            "required": { 
+                "frame_path": ("STRING", {"default": "C:/Users/Desktop",}), 
+                "fps": ("FLOAT", {
+                    "default": 30, 
+                    "min": 1,
+                    "max": 120,
+                    "step": 1,
+                    "display": "number"
+                }),
+                "video_name": ("STRING", {"default": "new_video"}),
+                "output_path": ("STRING", {"default": "./output"}),
+            },
+            "optional":{
+                "audio_path":("STRING",{"default": "C:/Users/audio.mp3",}),
+                }
+        }
+
+    RETURN_TYPES = ("STRING","STRING",)
+    RETURN_NAMES = ("frame_path","output_path",)
+    FUNCTION = "frames2video" 
+    OUTPUT_NODE = True
+    CATEGORY = "FFmpeg" 
+
+    def frames2video(self,frame_path,fps,video_name,output_path,audio_path=None):
+        try:
+            if not os.path.exists(frame_path):
+                raise ValueError("frame_path不存在（frame_path does not exist）")
+            
+            #判断output_path是否是一个目录
+            if not os.path.isdir(output_path):
+                raise ValueError("output_path不是目录（output_path is not a directory）")
+            output_path =  f"{output_path}\{video_name}.mp4" # 将输出目录和输出文件名合并为一个输出路径
+            # 获取输入目录中的所有图像文件
+            frame_path = Path(frame_path)
+            image_files = sorted(frame_path.glob('*.png')) + sorted(frame_path.glob('*.PNG')) + sorted(frame_path.glob('*.Png'))
+            
+            # 获取第一张图像的文件名,不要路径
+            image_name = os.path.basename(image_files[0])
+            print(f"第一张图像文件名: {image_name}")
+
+            # 获取图像文件名中的数字个数，包括零
+            num_digits = sum(c.isdigit() for c in image_name)
+            print(f"转换后的图像文件名: {num_digits}")
+
+            # 将图像文件名中的数字替换为%0{num_digits}d，只影响数字，不影响其他字符
+            image_name = re.sub(r'\d+', '%0{}d'.format(num_digits), image_name)
+            print(f"转换后的图像文件名: {image_name}")
+
+            # 构建输出目录
+            if not image_files:
+                print(f"Files in directory: {os.listdir(frame_path)}")
+                raise FileNotFoundError(f"No image files found in directory: {frame_path}")
+
+            # 获取第一张图像的尺寸
+            with Image.open(image_files[0]) as img:
+                width, height = img.size
+                scale = f'{width}:{height}'
+
+            if audio_path:
+                # 有音频文件，构建ffmpeg命令
+                ffmpeg_cmd = [
+                    'ffmpeg',
+                    '-framerate', str(fps),
+                    '-i', f'{frame_path}/{image_name}',
+                    '-i', audio_path,  # 添加音频文件路径
+                    '-vf', f'scale={scale}',
+                    '-c:v', 'libx264',
+                    '-preset', 'medium',
+                    '-crf', '28',
+                    '-pix_fmt', 'yuv420p',
+                    '-c:a', 'aac',  # 指定音频编解码器
+                    '-shortest',  
+                    '-y',
+                    str(output_path)
+                ]
+                # 执行ffmpeg命令
+                process = subprocess.Popen(ffmpeg_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+                stdout, stderr = process.communicate()
+                if process.returncode != 0:
+                    print(f"ffmpeg 执行失败，错误信息: {stderr.decode('utf-8')}")
+                else:
+                    print(f"视频合成成功: {output_path}")
+
+            else:
+                # 无音频文件，构建ffmpeg命令
+                ffmpeg_cmd = [
+                    'ffmpeg',
+                    '-framerate', str(fps),
+                    '-i', f'{frame_path}/{image_name}',
+                    '-vf', f'scale={scale}',
+                    '-c:v', 'libx264',
+                    '-preset', 'medium',
+                    '-crf', '28',
+                    '-pix_fmt', 'yuv420p',
+                    '-y',
+                    str(output_path)
+                ]
+                # 执行ffmpeg命令
+                process = subprocess.Popen(ffmpeg_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+                stdout, stderr = process.communicate()
+                if process.returncode != 0:
+                    print(f"ffmpeg 执行失败，错误信息: {stderr.decode('utf-8')}")
+                else:
+                    print(f"视频合成成功: {output_path}")
+            frame_path = str(frame_path) # 输出路径为字符串
+            return (frame_path,output_path)
+        except Exception as e:
+            raise ValueError(e)

nodes/video2frames.py

@@ -0,0 +1,121 @@
+import json
+import math
+import os
+import subprocess
+
+class Video2Frames:
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": { 
+                "video_path": ("STRING", {"default":"C:/Users/Desktop/video.mp4",}),
+                "output_path": ("STRING", {"default":"C:/Users/Desktop/output",}),
+                "frames_max_width":("INT", {"default": 0, "min": 0, "max": 1920}),
+            },
+        }
+
+    RETURN_TYPES = ("STRING", "FLOAT", "STRING", "INT","STRING")
+    RETURN_NAMES = ("frame_path", "fps", "audio_path", "total_frames","output_path")
+    FUNCTION = "video2frames"
+    OUTPUT_NODE = True
+    CATEGORY = "FFmpeg"
+  
+    def video2frames(self, video_path, output_path, frames_max_width):
+        try:
+            # 视频不存在
+            if not video_path.lower().endswith(('.mp4', '.avi', '.mov', '.mkv','.rmvb')):
+                raise ValueError("video_path不是视频文件（video_path is not a video file）")
+            
+            if not os.path.exists(video_path):
+                raise ValueError("video_path不存在（video_path does not exist）")
+            
+            #判断output_path是否是一个目录
+            if not os.path.isdir(output_path):
+                raise ValueError("output_path不是目录（output_path is not a directory）")
+            
+            # 判断frames_max_width是否是一个整数
+            if not isinstance(frames_max_width, int):
+                raise ValueError("frames_max_width不是整数（frames_max_width is not an integer）")
+            
+            # 提取音频
+            audio_path = os.path.join(output_path, 'audio.mp3')
+            video_path = os.path.abspath(video_path).replace("\ufeff", "").strip()
+            audio_cmd = [
+                'ffmpeg', '-i', video_path, 
+                '-q:a', '0', '-map', 'a', '-y', audio_path
+            ]
+            subprocess.run(audio_cmd)
+
+            # 获取视频帧率、时长、宽高信息
+            command = [
+                'ffprobe', '-v', 'error', '-select_streams', 'v:0', '-show_entries', 
+                'stream=avg_frame_rate,duration,width,height', '-of', 'json', video_path
+            ]
+
+            # 运行ffprobe命令
+            result = subprocess.run(command, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
+            # 将输出转化为字符串
+            output = result.stdout.decode('utf-8').strip()
+            print(output)
+            data = json.loads(output)
+            # 查找视频流信息
+            if 'streams' in data and len(data['streams']) > 0:
+                stream = data['streams'][0]  # 获取第一个视频流
+                fps = stream.get('avg_frame_rate')
+                if fps is not None:
+                    # 帧率可能是一个分数形式的字符串，例如 "30/1" 或 "20.233000"
+                    if '/' in fps:
+                        num, denom = map(int, fps.split('/'))
+                        fps = num / denom
+                    else:
+                        fps = float(fps)  # 直接转换为浮点数
+                else:
+                    raise ValueError("无法获取视频的帧率")
+                width = int(stream.get('width'))
+                height = int(stream.get('height'))
+                duration = float(stream.get('duration'))
+            else:
+                raise ValueError("无法获取视频信息")
+
+            # 计算总帧数
+            total_frames = math.ceil(fps * duration)
+            print(f"视频的帧率是: {fps}, 宽度是: {width}, 高度是: {height}, 时长是: {duration}, 总帧数是: {total_frames}")
+            # 提取帧
+            frame_path = os.path.join(output_path, 'frames')
+            os.makedirs(frame_path, exist_ok=True) # exist_ok=True表示如果目录已经存在，不会引发异常
+
+            # 计算输出宽度和高度以保持比例
+            if frames_max_width > 0:
+                if width > frames_max_width:
+                    out_width = frames_max_width
+                    out_height = int(height * frames_max_width / width)  # 按比例计算新高度
+                else:
+                    out_width = width
+                    out_height = height
+            else:
+                out_width = width
+                out_height = height
+                
+            command = [
+                'ffmpeg', '-i', video_path,  # 输入视频路径
+                '-vf', f'scale={out_width}:{out_height}',  # 使用scale滤镜缩放帧
+                os.path.join(frame_path, 'frame_%06d.png')  # 输出帧路径
+            ]
+            
+            # 执行命令并检查错误
+            if subprocess.run(command).returncode != 0:
+                print("生成序列帧时出错，请检查输入文件和路径。")
+
+            # 打印信息
+            print(f"序列帧图片输出路径: {frame_path}")
+            print(f"音频输出路径: {audio_path}")   
+            print(f"总帧数: {total_frames}")
+            print(f"帧率: {fps}")
+            print(f"输出路径: {output_path}")
+
+            return (frame_path, fps, audio_path, total_frames,output_path)
+        except Exception as e:
+            raise ValueError(e)

nodes_map.py

@@ -0,0 +1,19 @@
+from .nodes.addTextWatermark import *
+from .nodes.frames2video import *
+from .nodes.video2frames import *
+from .nodes.addImgWatermark import *
+
+NODE_CLASS_MAPPINGS = {
+    "Video2Frames": Video2Frames,
+    "Frames2Video": Frames2Video,
+    "AddTextWatermark": AddTextWatermark,
+    "AddImgWatermark": AddImgWatermark,
+}
+
+# A dictionary that contains the friendly/humanly readable titles for the nodes
+NODE_DISPLAY_NAME_MAPPINGS = {
+    "Video2Frames": "🔥Video2Frames",
+    "Frames2Video": "🔥Frames2Video",
+    "AddTextWatermark": "🔥AddTextWatermark",
+    "AddImgWatermark": "🔥AddImgWatermark",
+}

requirements.txt

@@ -0,0 +1 @@
+pathlib