Interesting to see something like this!

My computer science masters thesis was based on the same goal. I used a 2D convolution which meant you can merge images with inexact overlaps. I had to run a high-pass filter first to limit the image details to their edges only or else the convolution incorrectly matched bright areas.

In reality merging pictures is further complicated because the source images may be slightly rotated relative to each other and also due to the images being slightly curved due to lens distortion.

My supervisor wanted me to do a PHD on the topic!

What are the practical applications for this tool? Typically stitching images for something like panoramas requires significantly more advanced image processing algorithms because the pixels do not perfectly overlap.

Interesting! The example shows two images that appear to have a pixel-perfect matching region. Is that a requirement or does it work with images that are only somewhat similar?

Nice project of yours! I am a data science student but I never looked into Computer Vision. Until a few days ago, when I started watching a series of short courses on a YouTube channel called First Principles of Computer Vision [0]. I found it fascinating and the math behind is truly beautiful, concise and efficient.

[0] https://www.youtube.com/@firstprinciplesofcomputerv3258 strongly recommend to check-out any playlist. Best courses I have had since a long time.

look at those for loops! should look into fft-based correlation, can even do so with melon transform for scale and circular harmonic transform for rotation

I’ve been looking for something like this for creating surveys using drone footage - extract every “n” frames from the video, then stitch ‘em up somehow to make a “layer”.

There’s existing software for this kind of work, but I’ve been in the mood to reinvent the wheel a bit for some strange reason.

Nice work!

What's the latest state of the art in image stitching these days? From what I can tell, there was a bunch of research done on it in the past, but with all the recent advancements in AI, not much has changed on this front. I'd love to be wrong though!

Related to this, is there a name for the effect when you stitch together video frames into a static background while keeping the moving objects moving? The best example I can think of is this Bigfoot video[0, 1], where the shaky footage has been combined into a bigger canvas with "Bigfoot" moving through it. It's a combination of video stabilization and image panorama, but with some smarts to only keep one version of the moving object in each finished frame.

[0] https://www.youtube.com/watch?v=Q60mSMmhTZU [1] https://x.com/rowancheung/status/1641519493447819268

Nicely done and keep up the practice. I recall during my Masters needing to translate facial landmark points from a Cartesian coordinate system into points that could would appear on their respective images. It wasn't for anything major, I just wanted a visual representation of my work. Its these little "neat" projects that help build larger breakthroughs.

See also: Hugin - Panorama photo stitcher. I used to use a lot back in ~2006 for making panoramas. It automatically finds "control points" in your photos, figures out which ones are shared between the photos and uses that information to determine the relative positions of the photos, and your lens parameters.

Once it does that it can stitch the photos together. It does this by projecting the photos onto a sphere, and then taking a picture of that sphere using whatever lens parameters you want.

https://hugin.sourceforge.io/

Why a naive pixel matching library got so many likes here?

Oh shit thank you so much. I’ve always had to use photoshop for this

In the biomedical sciences, we typically use a tool called BigStitcher [0], which is bundled with ImageJ [1]

[0] https://www.nature.com/articles/s41592-019-0501-0 [1] https://imagej.net/plugins/bigstitcher/