image VS opencv-rust

From https://github.com/image-rs/image

And for image resizing I used the "image" crate: https://crates.io/crates/image. Note that while in development mode resizing an image takes a long time (several seconds for me), but in production mode it is lightning fast. I spent some time trying to figure this out :D

Why not just add support for QOI to image?

The image create is actually pretty good. It decodes a lot of stuff natively, and encodes to JPEG and PNG. I actually used it to make something very similar.

the image crate might be just what you're looking for. more specifically, the gif encoder can do exactly what you want.

Use image to read individual images and gif to compose them into an animation.

I've used fits-rs before, but it lacks interop with the awesome image crate, so I only really used it for quick prototyping and then rolled my own fits parser.

I would recommend using the image crate for any image manipulation.

https://github.com/image-rs/image has decoding support for WebP luma channel only. Could really use chroma decoding. There is a draft PR, but it's stalled.

Supports different image formats on any Android device.

There are rust language bindings for opencv. https://github.com/twistedfall/opencv-rust

That would seem to indicate you do not actually have OpenCV installed, then. I'd recommend perhaps following the steps for getting the Windows prerequisites set up that the OpenCV Rust bindings provide in their readme.

I added opencv-rust as a dependency to a project. It makes my builds and even cargo check s really slow. It seems to invoke the build.rs script every time even when just type checking. Is this intentional? Is there any way to avoid this?