James Adams
For configuring the training of some TensorFlow SSD models it's necessary to know the number of examples in the evaluation dataset. In order to come up with this number, we...
Add the ability to relabel Darknet (YOLO) annotation files. This is essentially a matter of replacing the label index, which is the first field in the annotation lines of the...
Use the [Codecov analysis](https://codecov.io/gh/monocongo/cvdata/branches) to locate areas where increased coverage is possible and add more test cases into the test suite to improve the score (above 90% is ideal).
The `openimages` module currently only supports getting object detection annotations with bounding boxes. Add the ability to also pull a dataset including object segmentation annotations.
Use [this script](https://github.com/tensorflow/models/blob/master/research/object_detection/dataset_tools/create_oid_tf_record.py) as a basis for this dataset creation process.
Clean up the README for clarity, in the style of [coco_assistant](https://pypi.org/project/coco-assistant/).
Darknet format annotations are not yet supported for resizing, making it necessary to first convert a dataset to be resized into one of the supported formats (KITTI or PASCAL) for...
Add a module with CLI for fetching images using the [Google Images Download](https://github.com/hardikvasa/google-images-download) package. We can possibly create corresponding annotations using a pre-trained object detection model such as YOLOv3 etc....
Follow the steps laid out [here](https://circleci.com/docs/2.0/language-python/).
Add a recipe script for creating a dataset including people (person image class), vehicles (Car and Truck image classes), and weapons (Handgun and Rifle and Shotgun label classes) from OpenImages...