torchvision_detection is a package of utility functions for performing object detection using torchvision models. The functions included in this package are mainly compiled from several sources into one common repository for easy access and use. These functions were aggregated directly from the sources below.

These functions and scripts are not part of the torch package, and were obtained from: https://github.com/pytorch/vision/tree/v0.12.0/references/detection

These functions were obtained from the GitHub repository of Modern Computer Vision with PyTorch by V Kishore Ayyadevara and Yeshwanth Reddy: https://github.com/PacktPublishing/Modern-Computer-Vision-with-PyTorch/blob/master/Chapter08/Training_Faster_RCNN.ipynb

pip install git+https://github.com/natsunoyuki/torchvision_detection

import detection
import fasterrcnn_utils

# Coco dataset.
dataset = detection.coco_utils.get_coco(root, image_set, transform, mode = "instances")

# Fasterrcnn model.
model = fasterrcnn_utils.model.fasterrcnn(num_classes = 2)

• https://pytorch.org/vision/stable/training_references.html
• https://github.com/pytorch/vision/tree/v0.12.0/references/detection
• https://github.com/PacktPublishing/Modern-Computer-Vision-with-PyTorch