PIQA v1.3.0

💥 Breaking news

PIQA 1.3.0 drops its custom complex module in favor of torch.complex, which is stable since PyTorch 1.12. Accordingly, the support for torch <= 1.11 has been dropped.

✨ What's new

• New Fréchet Inception Distance (FID) metric (8950f5e)
• All metrics now support PyTorch's tracing (d233ef8)
• The documentation has been refreshed and moved to Read the Docs (d233ef8, d670de1)
• New contributing guidelines (7ebd950)

🐛 Bug fixes

• Fix NaNs in LPIPS (#33) by @ilpoli

Full Changelog: v1.2.2...v1.3.0

PIQA v1.2.2

📝 Documentation

The documentation has been refactored and is now generated using Sphinx and Furo. Check it out at francois-rozet.github.io/piqa/.

⚡️ Improvements

• Replace torch.linalg.norm by l2_norm (48ec8c4)
• New function to disable debugging (ffc07e0)

🐛 Bug fixes

• Fix RuntimeError while using conv1d for 2 or 3-d convolutions with PyTorch 1.11 (7a56439)

PIQA v1.1.7

⬆️ Dependencies

PIQA 1.1.7 requires the torch.fft module released with PyTorch 1.8.0.

✨ Metrics

• VSI Visual Saliency-based Index (0125e90)
• FSIM Feature Similarity (0125e90)

⚡️ Improvements

• Make padding optional in SSIM (76415a9)
• Make downsampling optional in FSIM, GMSD, HaarPSI, MDSI and VSI (9fedf9f)

🐛 Bug fixes

• Fix JITting for utils.complex submodule (d098f4d)

PIQA v1.1.3

🚸 Type assertions

Object-oriented components (PSNR, SSIM, ...) now use type assertions to raise meaningful error messages. See the "Assert" section in the README for more information.

♻️ Refactor

In this release, the API has been heavily redesigned and simplified.

• The "user-friendly" functionals to compute the metrics were dropped in favor of object-oriented metrics (5994e34)
• Importing piqa now directly gives access to the object-oriented metrics (5994e34)

v1.1.0	v1.1.3
>>> from piqa import ssim >>> criterion = ssim.SSIM() >>> criterion(x, y) tensor(...)	>>> import piqa >>> criterion = piqa.SSIM() >>> criterion(x, y) tensor(...)

• The piqa.utils module was divided in three sub-modules (5994e34)
  1. piqa.utils.functional regroups the convolution and kernel helpers
  2. piqa.utils.complex is an API to manipulate "factice" complex tensors
  3. piqa.utils.color handles the color-space conversions
• JITing is now an option which can be enabled/disabled (f467070)
• Custom normalization functionals (tensor_norm & normalize_tensor) were dropped in favor of torch.linalg.norm (f744007)

PIQA v1.1.0

PIQA now has an official logo and a banner 🎉

⚡️ Improvements

• Update core components to support PyTorch JIT (c8ac2c3)
• Optimize Gaussian blur with separable convolutions (c8ac2c3)
• Simplify some components (ba07335, 0b5a85e, 785b449)

🐛 Bug fixes

• Fix MDSI automatic differentiation (f73b13c, c6d924d)

📝 Documentation

• Add table of available metrics in README
• Refresh the full documentation (types, shapes, etc.)
• Add mathematical definitions in docstrings

⬆️ Dependencies

PIQA 1.1.0 requires PyTorch 1.7.0 or above.

PIQA v1.0.7

✨ Metrics

• MS-GMSD Multi-Scale Gradient Magnitude Similarity Deviation (2364fbc)
• HaarPSI Haar Perceptual Similarity Index (3227b26)

⚡️ Improvements

• Buffer convolution kernels in object-oriented metrics (e6179ae)

🐛 Bug Fixes

• Fix inaccurate lpips.LPIPS results (43ae9d7)
• Fix RuntimeError in mdsi.mdsi on CUDA (0e89000)

PIQA v1.0.5

🚑 Critical fixes

• Pin correct dependencies (3860c5c)
• Fix FileNotFoundError when loading lpips.LPIPS weights (200460a)
• Fix dimension mismatch in tv.tv (fe983b6)

✅ Unit tests

Examples were added in the docstrings for a better user experience. They also acts as unit tests thanks to the doctest and unittest libraries.

PIQA v1.0.0

📦 Package

• The package is renamed to PyTorch Image Quality Assessment (PIQA)
• PIQA is deployed on PyPi

✨ Metrics

• GMSD Gradient Magnitude Similarity Deviation (8c1e1dc)
• MDSI Mean Deviation Similarity Index (e79b340)

⚡️ Improvements

• Use cascading named arguments (**kwargs) to simplify function signatures (3d1fd4c)
• Replace duplicated reduction if elif else statement by a function (c9719b2)
• ...

📝 Documentation

• Update code style to be compliant with the Google Python Style
• Update docstrings for consistency with PyTorch docs

SPIQ v0.0.2-beta

First release of the Simple PyTorch Image Quality (SPIQ) package 🎉

✨ Metrics

• PSNR Peak Signal-to-Noise Ratio (d79e5f3)
• SSIM Structural Similarity (c1addd2)
• MS-SSIM Multi-Scale Structural Similarity (be1821f)
• TV Total Variation (2be72f6)
• LPIPS Learned Perceptual Image Patch Similarity (6a17031)

📝 Documentation

• Installation instructions and snippet for getting started in README
• Documentation fully contained in the components' docstrings
• Documentation website generated automatically from the docstrings using pdoc

📄 License

This package is under the MIT License.