F-G Fernandez

Results 11 issues of F-G Fernandez

This issue is to be used to track the roadmap of docTR for release v0.6.0, and collect feedback from users & contributors. For the v0.5.2 roadmap, please see #967 -...

critical

Most users of the library are more interested in existing pretrained models to use for inference rather than training. For this reason, it's important to ensure we can easily export...

critical
module: models
framework: tensorflow

Low-power devices such as Raspberry PIs are widely used by developers to come up with exciting products. Adding customized builds for ARM architectures would greatly help their efforts!

topic: build
side-project
topic: arm

This PR introduces the following modifications: - transforms: updated the input and output signature of `RandomRotate` - character classification: expanded data augmentations for PyTorch - obj detection: switched StepLR &...

type: enhancement
ext: references
module: transforms
topic: text recognition
topic: character classification
topic: object detection

Unfortunately, one of the project dependencies does not have any conda release or any way to make one. I opened an issue on their repo https://github.com/pymupdf/PyMuPDF/issues/938 to track this, but...

type: bug
topic: build

The current pretrained artefact detection model was trained on a fully synthetic dataset. While this comes with several advantages, the dataset has a distribution that is still a bit far...

type: enhancement
module: models
framework: pytorch
topic: object detection

The library doesn't have clear information on the consequences of image transformation using different framework backends. Some need to be investigated: - appearance of artefacts during interpolation with some methods...

help wanted
ext: references
module: transforms
framework: pytorch
framework: tensorflow

As discussed in #654, the artefact detection needs to improve its robustness. In order to do so and prevent overfitting, I would suggest gradually extending the list of our supported...

module: transforms
topic: object detection

Add a `doctr.models.utils` module to compress existing models and improve their latency / memory load for inference purposes on CPU. Some interesting leads to investigate: - [x] FP conversion (#10)...

help wanted
module: models
framework: pytorch
framework: tensorflow

Currently, as specified in #609, the API template only supports single image input. With the latest version of docTR, it would be quite easy to change this to support PDF...

type: enhancement
ext: api