Generate a video with an image every 5 seconds using ffmpeg

To generate a video where a single image is displayed for five-second intervals using ffmpeg, you must loop the input image and set the output frame rate to 1/5. This approach ensures that each frame represents exactly five seconds of duration, creating a slideshow effect without requiring complex filtering or multiple input files.

Understanding the Input Loop Mechanism

The core challenge in this task is managing time. Standard images are static; they do not have a duration. When ffmpeg processes an image as an input, it typically treats it as a single frame with a duration of zero unless explicitly told otherwise. To make that single frame last for five seconds, we need to manipulate how ffmpeg interprets the input stream's timing. We achieve this by combining two specific flags: -loop 1 and -framerate 0.2. The -loop 1 flag tells ffmpeg to read the image repeatedly until the end of the input is reached. Without this, ffmpeg would process the image once and then stop, resulting in a video of negligible length. However, simply looping is not enough because ffmpeg needs to know how fast to play back those loops. By setting the framerate to 0.2, we are defining that only one frame should be output every five seconds (since 1 divided by 0.2 equals 5). This effectively stretches the single frame over a five-second period. It is crucial to place these options before the input file specification so that ffmpeg applies them to the image source rather than the final output container.

Why Not Use -t Duration?

A common misconception is to use the -t flag to limit the total duration. While -t 10 would create a ten-second video, it does not inherently control the interval between distinct visual changes if you were trying to combine multiple images. For a single image repeated, the combination of loop and framerate is the most robust method because it defines the temporal resolution of the stream itself. If you were to use a different framerate, such as 25, the video would play through the looped image extremely quickly, making it appear as a flash rather than a sustained display.

Constructing the Command

The command structure follows a logical sequence: global options, input definition, filter chain (if necessary), and output settings. Below is the standard command template for generating a video from a single image with a five-second interval.

ffmpeg -loop 1 -i input.jpg -vf "fps=1/5" -c:v libx264 -pix_fmt yuv420p output.mp4

Let us break down each component to ensure you understand the underlying mechanics.

• -loop 1: This instructs ffmpeg to loop the input image indefinitely. It prevents the encoder from exiting immediately after processing the first frame.
• -i input.jpg: This specifies the source image file. You can replace input.jpg with any supported image format, such as PNG or TIFF.
• -vf "fps=1/5": This is the video filter that sets the output frame rate. By using the fraction 1/5, we explicitly request one frame every five seconds. This is mathematically equivalent to 0.2 fps.
• -c:v libx264: This selects the H.264 video codec, which is widely compatible with web browsers and media players. You can change this to mpeg4 or libvpx-vp9 depending on your requirements.
• -pix_fmt yuv420p: This ensures pixel compatibility. Many older players and web standards require the YUV 4:2:0 chroma subsampling format to render colors correctly.
• output.mp4: The destination filename. You can choose other extensions like .mkv or .avi if needed.

Alternative Approaches and Considerations

While the fps filter is the most direct method, some users prefer using the -r flag directly in the output options. The command would look like this:

ffmpeg -loop 1 -i input.jpg -r 0.2 -c:v libx264 -pix_fmt yuv420p output.mp4

Both methods produce identical results. The -r flag sets the output framerate, while the -vf fps=... filter performs the same operation within the filter graph. The choice between them is largely stylistic. However, using the filter graph offers more flexibility if you plan to add additional transformations later, such as scaling or adding text overlays. If you need to combine multiple images, each lasting five seconds, you cannot rely on a simple loop. Instead, you would need to concatenate individual segments or use a complex filter graph that maps each image to a specific duration. For single-image scenarios, the loop method remains the most efficient and least resource-intensive solution. For users encountering issues with output formats, particularly when dealing with hardware acceleration or specific container constraints, refer to our guide on Troubleshooting: 'v4l2' is not a suitable output format to resolve potential encoding errors. Additionally, if you are unsure whether your generated video contains the expected streams, you can verify the stream composition by following the steps in How to detect with ffmpeg if the stream has video or audio or both.

Verifying the Output

After running the command, it is essential to verify that the duration matches your expectations. You can use ffprobe to inspect the metadata of the generated file. Run the following command:

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

This will return the total duration in seconds. Since the video consists of a single frame repeated at a rate of 0.2 fps, the duration might be very short if ffmpeg optimizes away duplicate frames. To ensure the video plays for a specific total time, you may need to add a -t flag to specify the total length, or adjust the loop logic if you intend for the video to run for minutes rather than seconds. If you find that your video has unexpected resolution or progressive scan issues, check out our troubleshooting article on Troubleshooting FFmpeg-created HLS stream showing 0p (zero lines, progressive) resolution option to ensure your output parameters are correctly aligned with your playback environment.

Frequently Asked Questions

Can I use PNG images instead of JPG?

Yes, ffmpeg supports PNG inputs just as well as JPG. Simply replace input.jpg with input.png in the command. Note that PNG files are larger, which may increase processing time slightly but preserves transparency if you use alpha channels.

How do I add audio to this video?

You can add an audio track by including another input file with the -i flag before the video input. For example, -i audio.mp3 followed by the video input. Then use the -shortest flag to trim the video to match the audio length, or let the video loop continuously while the audio plays.

What if my video plays too fast?

If the video plays too fast, it means the framerate is too high. Ensure you are using fps=1/5 or -r 0.2. A value of 1 would result in one frame per second, making the image change every second, while 0.2 makes it change every five seconds.

Does this method work for batch processing multiple images?

No, this specific command uses -loop 1 which repeats a single image. For batch processing multiple images with different durations, you would need to create a list file or use a script to generate individual commands for each image, then concatenate the resulting videos.

Why is my output file size very small?

A small file size is normal for a video consisting of identical frames. Since there is little motion data for the encoder to compress, the resulting file will be minimal. If you need a longer video, you must specify a total duration using the -t flag, as the loop alone does not define an end time.