The Lepton compression library is designed for lossless compression of baseline and progressive JPEGs up to 22%, with exact bit-by-bit recovery of the original JPEG. The primary use case is for storing JPEGs in a cloud-storage system. Metadata headers, and even invalid content is preserved as-is.
This is a port of the C++ Lepton JPEG compression tool that was released by DropBox in this location: dropbox/lepton: Lepton is a tool and file format for losslessly compressing JPEGs by an average of 22%. (github.com)
Due to the work involved in doing a complete security audit on the C++ code, and the fact that DropBox has deprecated the codebase, we created a port of the library to Rust, which has almost identical performance characteristics with the advantage of all the safety features the Rust offers.
The source of the library itself is under the src directory, with integration tests in the test directory. There are various test images under the images folder.
Building the project is fairly straightforward if you have Rust 1.65 or later installed (older version will warn about unstable features such as scoped threads). cargo build and cargo test do what you would expect, and cargo build --release creates the optimized release version.
Some operations are vectorized such as the IDCT using the Wide crate, so you can get a significant boost if you enable +AVX2.
There is an lepton_jpeg_util.exe wrapper that is built as part of the project. It can be used to compress/decompress and also to verify the test end-to-end on a given JPEG. If the input file has a .jpg extension, it will encode. If the input file has a .lep extension, it will decode back to the original.jpg.
It supports the following options:
lepton_jpeg_util.exe [options] <inputfile> [<outputfile>]
| Option | Description |
|---|---|
-threads:n |
Runs with a maximum of n threads. For encoding, this limits the amount of parallelism that can be gotten out of the decoder. |
-dump |
Dumps the contents of a JPG or LEP file, with the -all option, it will also dump the cooefficient image blocks |
-noprogressive |
Will cause an error if we encounter a progressive file rather than trying to encode it |
-verify |
Reads, encodes and unencodes verifying that there is an exact match. No output file is specified. |
-iter:n |
Runs N iterations of the operation. Useful when we are running inside a profiler. |
Link to overall design of library
There are many ways in which you can participate in this project, for example:
- Submit bugs and feature requests, and help us verify as they are checked in
- Review source code changes or submit your own features as pull requests.
- The library uses only stable features, so if you want to take advantage of SIMD features such as AVX2, use the Wide crate (see the idct.rs as an example) rather than intrinsics.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.
Copyright (c) Microsoft Corporation. All rights reserved.
Licensed under the Apache 2.0 license.