Temporal Scale-Space Toolbox for Python.
Project description
pytempscsp : Temporal Scale Space Toolbox for Python
For performing temporal smoothing with the time-causal limit kernel and for computing discrete temporal derivative approximations by applying temporal difference operators to the smoothed data.
This code is the result of porting a subset of the routines in the Matlab package tempscsp to Python, however, with different interfaces for the functions.
For examples of how to apply these functions for smoothing temporal signals to different temporal scales in a fully time-causal manner, please see the enclosed Jupyter notebook tempscspdemo.ipynb.
For more technical descriptions about the respective functions, please see the documentation strings for the respective functions in the source code in tempscsp.py.
Installation
This package is available through pip and can installed by
pip install pytempscsp
This package can also be directly downloaded from GitHub:
git clone git@github.com:tonylindeberg/pytempscsp.git
References
Lindeberg (2023) "A time-causal and time-recursive temporal scale-space representation of temporal signals and past time", Biological Cybernetics 117 (1-2): 21-59. (Open Access)
Lindeberg (2016) "Time-causal and time-recursive spatio-temporal receptive fields", Journal of Mathematical Imaging and Vision 55(1): 50-88. (Open Access)
The time-causal limit kernel was first defined in Lindeberg (2016), however, then also in combination with a spatial domain, and experimentally tested on video data. The later overview paper (Lindeberg 2023) gives a dedicated treatment for a purely temporal domain, and also with relations to Koenderink's scale-time kernels and the ex-Gaussian kernel.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for pytempscsp-0.9.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | fe36deb2960944b429eba0df0cb566de4ee15940f6be5c8c9ed06c69c9050001 |
|
MD5 | 87fba22da3ec1fb7c17bdd128b87d98f |
|
BLAKE2b-256 | 2fa6898a96d284f9de2d35012f88c7baf9d9239517fd0eaee57145054fd49c8a |