π Geometric Computer Vision Library for Spatial AI
Python artificial-intelligencecomputer-visiondeep-learninghacktoberfestimage-processingmachine-learningneural-networkpythonpytorchroboticsspatial-ai
Codec
=====
.. currentmodule:: kornia.enhance
A differentiable JPEG codec, useful to simulate compression artifacts inside a training loop.
It accepts RGB ``(*, 3, H, W)`` tensors in ``[0, 1]`` and a quality tensor of
shape ``(1,)`` or ``(N,)``, one per flattened image. An integer image raises
:exc:`~kornia.core.exceptions.TypeCheckError`; convert 8-bit data with ``image.float() / 255`` first.
.. autofunction:: jpeg_codec_differentiable
.. autoclass:: JPEGCodecDifferentiable