Single-photon single-particle tracking (SP²T) is the first offline single-particle tracking algorithm to directly track using 1-bit binary frames produced by single-photon detectors such as single-photon avalanche diodes (SPADs).
SP²T is written in Julia, so please download and install Julia (v1.11 or newer) following its instructions. Thanks to Julia's built-in package manager, most packages SP²T relies on do not require manual installation--they will be installed automatically when you install SP²T. The Installation section will list a few exceptions.
If it interests you, Project.toml lists the required Julia packages.
We write SP²T so that it can run on CPU or GPU. However, running on GPU is much (>10x) faster. For this reason, you are highly encouraged to try it out.
Although tested mostly on Nvidia GPUs, SP²T itself is not limited to any GPU company as long as the packages it relies on (e.g., NNlib.jl) support the GPU you have.
This is not a requirement; however, for the best experience, you are encouraged to use an integrated development environment that supports Julia. Visual Studio Code (VSCode) is the best choice: it is free and the best-supported platform by the Julia community. Please check how to install VSCode and its Julia extension here.
Once Julia, VSCode, and the Julia extension for VSCode are all installed. It is just one step away from getting SP²T installed. To run the commands in this section, please first open the Julia REPL in VSCode. You can do so by going to "Help>Show All Commands," then typing Julia: Start REPL and hitting Enter.
Now Julia REPL should be running, and we are ready to proceed!
SP²T is currently under active development and is still experimental, so it is not yet in Julia's official registry. However, you can try it by typing:
]add https://github.com/LabPresse/SP2T.jland hitting Enter in the Julia REPL.
Congrats, you have just installed SP²T! Although nothing else is required in principle, SP²T is designed to be minimal and only to contain the core functions for single-photon tracking. Therefore, it cannot visualize the results and save them to your hard disk. But don't worry; many well-written Julia packages can help us! We need two additional packages to run the example scripts, Distributions.jl and JLD2.jl. As they are in Julia's official registry, installing them is as easy as ]add Distributions, JLD2.
As mentioned in the Prerequisites, GPU is highly recommended. Please refer to the JuliaGPU website for information on what and how to install it based on your hardware.
Once SP²T is installed, you can download the example scripts in examples/2D. First, open simulation.jl in VSCode; running it will generate some example data. Then, you can open and run either inference_cpu.jl or inference_gpu.jl for a test calculation using the generated data on CPU or GPU, respectively.
The conceptual basis of SP²T is the same as the BNP-Track. The code is modified to cope with single-photon datasets.