Image to Video in Make

[Jeffrey_Brand]

What is your goal?

Dear Make community,

Since I was helped extremely well here last week, I thought: let’s just try again.

I’m currently working on building a content factory for Facebook and Instagram, with the intention of expanding this later to other platforms. I’m building this together with ChatGPT. However, ChatGPT has its limitations—especially when it comes to coding.

Because Make is completely new to me and this is purely a hobby project, I also have zero coding knowledge.

What I’m currently running into in my scenario is the following:
It works perfectly fine in generating images, coming up with captions, and ultimately posting everything to social media via Buffer.

But after about two weeks, I’ve started to notice that when I upload content this way, I’m only reaching my own followers, while I actually want to reach non-followers as well (for promotional purposes).

Last week, I manually turned one image into a Reel (no music or anything), and suddenly it got 4x more views, with about 60% coming from people who don’t follow me.

So my idea was:
Convert the images that are uploaded in the current scenario into videos instead.

Unfortunately, this turned out to be a lot less simple than I hoped.

What I’ve tried so far:

Using Cloudinary.
When I change the JPG to MP4 in the URL, I do get a video.
But when I add du_5 to the URL, the video does not become 5 seconds long.
I also tried replacing image with video in the URL, but that didn’t work at all.

Cloudinary seems to be one of the more affordable solutions, but every small mistake in the URL is very unforgiving.

After Cloudinary didn’t work out, ChatGPT suggested Replicate.
I added an API key and everything, but since Replicate doesn’t have a Make module, I’d have to use it via an HTTP request.
This resulted in an extremely long and complex piece of code, which is even less desirable for me.

So my actual question is:

Can someone help me set up a correct Cloudinary URL to turn an image into a video of a fixed duration?

Or does anyone know a different approach to solve this problem?

Here is the shared scenario link:

Probeersel (hybride) content factory - Make.com Automation Scenario

For clarity:
I’m currently working in the lower “photo” route. After Dropbox, I would like to turn the image into a video, so it can then be sent separately to the platforms.

There will be another adjustment later:

On the Facebook side, Buffer will be removed and posting will be fully automatic.

On the Instagram side, it will remain semi-automatic, since videos cannot be scheduled there.

Thanks in advance!

[Jeffrey_Brand]

Well, i have to say that google AI helped me better with this question…
Because somehow i got this now:

https://res.cloudinary.com/du5hcoyph/image/upload/c_fill,ar_9:16,e_zoompan:du_10;from_(zoom_1.0);to_(zoom_1.0)/v1770498698/xgc8ghqa3unr6a7gjuqq.mp4.

The only thing missing according to google is de g_auto but if i implement this. The link shuts down:
https://res.cloudinary.com/du5hcoyph/image/upload/c_fill,ar_9:16,g_auto/e_zoompan:du_10;from_(zoom_1.0);to_(zoom_1.0)/v1770498698/xgc8ghqa3unr6a7gjuqq.mp4.

Anyhow, if someone knows a better method i’m open for that!

[damato]

Hi @Jeffrey_Brand

Don’t you use Cloudinary “Transform a Resource” Make module?

Can you point to the specific ID of the module that is responsible for the transformation? Where does the input come from? Your scenario is huge.

Also, have you tried the Cloudinary Transformation Builder? You can use Converse mode to ask AI to build the transformation URL for you.

EDIT: Just checked their docs, and found out you can’t use automatic gravity with the panzoom effect. You need to specify where it’s coming from and where it’s going to. something like this:

https://res.cloudinary.com/du5hcoyph/image/upload/c_fill,ar_9:16,e_zoompan:du_10;from_(g_east;zoom_1.0);to_(g_west;zoom_2.0)/v1770498698/xgc8ghqa3unr6a7gjuqq.mp4

@damato

[Jeffrey_Brand]

To be completely honest: no, I’m currently using Upload a Resource. From that URL, I generate something.

I’m building this together with ChatGPT (whether this is actually the right approach… I guess I’ll find out later haha), but it explicitly told me not to use Transform a Resource.

Thanks for confirming that the auto function doesn’t work properly with pan & zoom.
The original idea was to get a “static” video. With this zoom setup it’s at least minimal, so that helps.

At the moment, everything works reasonably well. When uploading to Facebook, it correctly uploads as a video. However, even though the link looks like this:

https://res.cloudinary.com/du5hcoyph/image/upload/c_fill,w_1080,h_1920,e_zoompan:du_10;from_(zoom_1.0);to_(zoom_1.0)/v1770498698/{{212.public_id}}.mp4

or:

https://res.cloudinary.com/du5hcoyph/image/upload/c_fill,w_1080,h_1920,e_zoompan:du_10;from_(zoom_1.0);to_(zoom_1.0)/v1770498698/y4dnajgshomjf1ahfb34.mp4

I’m running into an issue where the transformation I’m requesting from Cloudinary via the URL seems to be too heavy. Because of that, Cloudinary outputs 720 × 1280 instead of 1080 × 1920.
This results in black sidebars in my Facebook post.

After some searching, it turns out this happens because the operation Cloudinary has to perform is too demanding. I then removed the entire zoompan transformation from the URL, and indeed — it correctly outputs 1080 × 1920 again.

The only problem is: without zoompan, it’s no longer a video.

So my actual question is:
Is there an alternative to zoompan that allows the image to stay (almost) completely static, but is less resource-intensive in the URL, so the output resolution remains 1080 × 1920?

P.S. The Transformation Builder looks really handy!

[Jeffrey_Brand]

First of all, my apologies — I might be a bit impatient — but I wanted to let you know what I’ve already tried myself, and that unfortunately it still hasn’t worked.

What I want to achieve:
I want to turn an image into a (static) video. This way, I can create a Reel on Instagram or Facebook instead of a regular post. Reels have a larger reach, which is what I’m aiming for.

What I’ve tried so far:

• I tried using Cloudinary “transform a resource.”

• I tried CloudConvert, but it also cannot convert a JPG or PNG into an MP4.

• I explored the Cloudinary transformation tool itself, but nowhere does it describe how to achieve what I’m trying to do.

• I tried using Cloudinary’s AI assistant, but unfortunately that didn’t help either.

• I continued experimenting by modifying the URL of my Cloudinary “upload a resource.”

So far, the zoompan transformation is the only thing that turns my image into a video. However, I’m still running into the issue that the video automatically gets cropped to 780 x 1280, while I need it to be 1080 x 1920. Otherwise, I get black bars on Facebook, and Instagram won’t allow me to post it properly.

When you ask AI, it says you need to change “image” to “video” in the URL. But the moment I do that, the URL immediately stops working… so unfortunately that’s not an option.

At this point, I feel like I’ve tried almost everything, and I can hardly believe there isn’t another way to make this work.

http://res.cloudinary.com/du5hcoyph/image/upload/v1770842438/rjjya0xwpda6dbwj1dos.jpg

Maybe I need to think completely differently than using Cloudinary or CloudConvert. If anyone has suggestions, I’d really appreciate it!

Once I get this working, I can finally set my scenario live.

Thanks in advance for the help!