1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
|
# embeddings-sort
This program can sort images such that ones with similar motives are close together. This is accomplished by using [AI](https://github.com/minimaxir/imgbeddings) to extract the meaning of the image, and then approximating a travelling-salesperson-tour through all of them.
As a bonus feature, this program can also sort the images by hue, brightness or color, though the results for this could be improved by using a less generalized algorithm.
The sorting can be accessed by letting the progam print the image paths in order, or by copying/symlinking the images into a new directory.
Detailed usage:
```
Usage: embeddings-sort [OPTIONS] [IMAGES]...
Arguments:
[IMAGES]...
Options:
-e, --embedder <EMBEDDER> Characteristic to sort by [default: content-euclidean] [possible values: brightness, hue, color, content-euclidean, content-angular-distance, content-manhatten]
-s, --symlink-dir <SYMLINK_DIR> Symlink the sorted images into this directory
-o, --copy-dir <COPY_DIR> Copy the sorted images into this directory. Uses COW when available
-c, --stdout Write sorted paths into stdout, one per line
-0, --stdout0 Write sorted paths into stdout, null-separated. Overrides -c
-b, --benchmark Output total tour length to stderr
--tsp-approx <TSP_APPROX> Algorithm for TSP approximation. Leave as default if unsure [default: christofides] [possible values: mst-dfs, christofides, christofides-refined]
-h, --help Print help
```
## Insides
The chrisofides implementation uses an approximated min-weight matching algorithm, which may be non-ideal, though I haven't benchmarked how much of a difference it makes (mainly due to the implementation complexity of an exact algorithm, which would also increase the implementations complexity from O(n²) to O(n³) where n is the number of given images).
christofides-refined is planned to be christofides but with an O(n²) 2-opt-swapping step added after the main algorithm. Implementing this efficiently will also require some algorithmic trickery, so it's not ready yet.
|
