* ResNet50 for Residual Networks, introduces a key innovation using residual blocks. This allows the training of
very deep networks without encountering the vanishing gradient problem. ResNet50 is primarily focused on image classification
and does not provide object localization. It can recognize objects from a vast set of more than 1,000 classes, covering a wide
range of objects, animals, and scenes. For specific details on these classes, you can refer to the file "classification_classes_ILSVRC2012.txt".

resnet50.onnx
classification_classes_ILSVRC2012.txt
resnet152-v2-7.onnx

Taken from: https://github.com/onnx/models/
Licensing : Apache-2

* EfficientnetB7 is an image classification model that works much better than ResNet-152. Reference: https://huggingface.co/google/efficientnet-b7/blob/main/README.md

efficientnetb7.onnx

Taken from: https://huggingface.co/google/efficientnet-b7
Licensing : Apache-2


* Models designed for object detection, capable of recognizing and extracting the location of objects within an image.
The limitation on the number of recognizable objects is set to 80. You can find the details of these objects in the file
"coco.names". YOLO nano is known for its speed, making it the fastest model, while YOLO XLarge provides more accurate
predictions at the expense of speed. Two versions are used: 5 and 11.

yolov5n_batch_16_s320.onnx
yolov5x_batch_16_s320.onnx
coco.names

Taken from: https://github.com/ultralytics/yolov5
Licensing : AGPL-3.0

yolo11n.onnx
yolo11x.onnx

Taken from: https://github.com/ultralytics/ultralytics
Licensing : AGPL-3.0