tensorflow/graphics
A TensorFlow library providing differentiable graphics layers like spatial transformers and renderers for neural network architectures.
Not currently ranked — collecting fresh signals.
star history
TensorFlow Graphics offers a collection of differentiable graphics layers that can be embedded into neural network architectures. It bridges computer graphics and deep learning by providing tools like spatial transformers, camera models, lighting models, and differentiable renderers. These layers enable networks to leverage geometric priors and constraints for more robust, efficient, and self-supervised training.
Frequently asked
- What is tensorflow/graphics?
- A TensorFlow library providing differentiable graphics layers like spatial transformers and renderers for neural network architectures.
- Is graphics open source?
- Yes — tensorflow/graphics is open source, released under the Apache-2.0 license.
- What language is graphics written in?
- tensorflow/graphics is primarily written in Python.
- How popular is graphics?
- tensorflow/graphics has 2.8k stars on GitHub.
- Where can I find graphics?
- tensorflow/graphics is on GitHub at https://github.com/tensorflow/graphics.