Skip to main content

Tool to work with annotation formats

Project description

pascal

Python utility to work with PascalVoc annotation format

Image examples from PascalVoc2007 dataset

Code Example
import json
from pathlib import Path

from PIL import Image

from pascal import annotation_from_xml
from pascal.utils import save_xml

ds_path = Path("/home/VOCtest_06-Nov-2007/VOCdevkit/VOC2007")

img_src = ds_path / "JPEGImages"
ann_src = ds_path / "Annotations"

out_labelme = ds_path / "converted_labelme"

attr_type_spec = {"truncated": bool, "difficult": bool}

label_map = {"car": 1, "dog": 0, "person": 2, "train": 3}

if __name__ == "__main__":
    for file in img_src.glob("*.jpg"):
        # Get annotation file path
        ann_file = (ann_src / file.name).with_suffix(".xml")
        # Make pascal annotation object
        ann = annotation_from_xml(ann_file, attr_type_spec)
        print(ann)
        # Save to xml file (same as ann_file)
        xml = ann.to_xml()
        out_xml_name = file.with_suffix(".xml").name
        save_xml(out_xml_name, xml)
        # Save yolo annotation
        yolo_ann = ann.to_yolo(label_map)
        out_yolo_name = file.with_suffix(".txt").name
        with open(out_yolo_name, "w") as f:
            f.write(yolo_ann)
        # Convert to labelme and save json file
        res = ann.to_labelme(file, save_img_data=False)
        with open((out_labelme / file.name).with_suffix(".json"), "w") as f:
            json.dump(res, f, indent=2)
        # Draw objects
        img = Image.open(file)
        draw_img = ann.draw_boxes(img)
        draw_img.show()
Visualization example:
draw_img = ann.draw_boxes(img)
draw_img.show()

vis_example

Installation

From source

python setup.py install

Using pip

pip install pascal-voc

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pascal_voc-2.1.8-py3-none-any.whl (218.3 kB view details)

Uploaded Python 3

File details

Details for the file pascal_voc-2.1.8-py3-none-any.whl.

File metadata

  • Download URL: pascal_voc-2.1.8-py3-none-any.whl
  • Upload date:
  • Size: 218.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.10.12

File hashes

Hashes for pascal_voc-2.1.8-py3-none-any.whl
Algorithm Hash digest
SHA256 e27655839a22cd263b4fe4c319c44bc2019f578cf51380fec19e25a7d1d4c12d
MD5 9a827acbca23713260d82aa65ff05864
BLAKE2b-256 8b3664a81c27fae4900f8459b9735e0eb5f731019059d4c864a71e57b78f3ff2

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page