如何用FFmpeg为三段拼接视频添加淡入淡出过渡效果

Hey there! Let's tackle this FFmpeg problem step by step—since you're not familiar with it, I'll break everything down clearly so you can follow along. The goal is to add a fade-out to the end of your opening video, a fade-in to the start of your closing video, and keep the transition smooth between all three clips.

This method lets you process all three videos and combine them in one command, which saves you from creating intermediate files. Here's the command you'll use, with explanations for each part:

ffmpeg -i 开场视频.mp4 -i 主视频.mp4 -i 结尾视频.mp4 -filter_complex \
"[0:v]fade=t=out:st=X:d=Y[v0]; \
 [0:a]afade=t=out:st=X:d=Y[a0]; \
 [2:v]fade=t=in:st=0:d=Y[v2]; \
 [2:a]afade=t=in:st=0:d=Y[a2]; \
 [v0][a0][1:v][1:a][v2][a2]concat=n=3:v=1:a=1[v][a]" \
-map "[v]" -map "[a]" -c:v libx264 -c:a aac output.mp4

关键参数解释： #

• X: The start time of the fade-out on your opening video. Calculate this as (duration of opening video) - Y. For example, if your opening video is 10 seconds long and you want a 2-second fade-out, X = 8.
• Y: The length of the fade effect (both fade-out and fade-in). 1-3 seconds is usually natural—adjust based on your preference.

How to get the exact duration of your opening video: #

Run this command in your terminal to get the precise length (it'll output a number like 10.000000):

ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 开场视频.mp4

What each filter does: #

• [0:v]fade=t=out:st=X:d=Y[v0]: Applies a video fade-out to the first (opening) clip, starting at X seconds and lasting Y seconds.
• [0:a]afade=t=out:st=X:d=Y[a0]: Syncs an audio fade-out with the video fade-out (critical for smooth transitions—don't skip this!).
• [2:v]fade=t=in:st=0:d=Y[v2]: Applies a video fade-in to the third (closing) clip, starting at 0 seconds (the very beginning) and lasting Y seconds.
• [2:a]afade=t=in:st=0:d=Y[a2]: Syncs an audio fade-in with the video fade-in.
• The concat filter combines all processed clips into one final video and audio stream.

If you prefer to test each step separately, you can process the opening and closing videos first, then combine all three using your original text file method:

1. Process the opening video (add fade-out):

ffmpeg -i 开场视频.mp4 -filter:v "fade=t=out:st=X:d=Y" -filter:a "afade=t=out:st=X:d=Y" 开场_处理后.mp4

2. Process the closing video (add fade-in):

ffmpeg -i 结尾视频.mp4 -filter:v "fade=t=in:st=0:d=Y" -filter:a "afade=t=in:st=0:d=Y" 结尾_处理后.mp4

3. Create a new text file (e.g., new_input.txt) with the processed files:

file '开场_处理后.mp4'
file '主视频.mp4'
file '结尾_处理后.mp4'

4. Combine the clips:

ffmpeg -f concat -safe 0 -i new_input.txt -c copy output.mp4

• Use -c copy to copy streams directly (fast!), but if you get an error about mismatched video/audio parameters, replace -c copy with -c:v libx264 -c:a aac to re-encode everything to match.

• Match video/audio parameters: Make sure all three clips have the same resolution, frame rate, pixel format, audio sample rate, and channel count. If not, the concat filter may fail. You can add a scale filter to the complex command to fix resolution mismatches (e.g., [1:v]scale=1920:1080[v1] to resize the main video to 1080p).
• Test with short clips: If your videos are long, test the command with trimmed versions first to avoid wasting time on full-length renders.
• Adjust fade timing: Play around with Y (fade duration) to find what feels natural for your content—1.5 seconds is a great starting point if you're unsure.

内容的提问来源于stack exchange，提问作者user7179690