Gain insights into using PyTorch for image classification and detection, delve into model implementation, and explore deployment on edge devices with ONNX and OpenVINO.

Image Classification and Object Detection using CNNs-01.png

pytorch.tar.gz

Basics_Of_CNN-ex1

Basics_Of_CNN-ex2

Basics_Of_CNN-vgg16

Basics_Of_CNN-inceptionv1

Basics_Of_CNN-residuals

Basics_Of_CNN-depthwise

Train-data

Train-training

Train-trainingFT

Train-transfer

deploy-onnx

deploy-openvino

deploy-comparison

prepare_dataset_openimages

train_yolov7

inference_yolov7

onnx_yolov7

Image classification and object detection have gained widespread use in recent years. Content categorization and monitoring, disease diagnosis from medical images, identifying terrain in satellite images, and detecting road elements for self-driving cars are classification problems at their core. PyTorch is a popular framework for these tasks—offering a useful mix of user-friendliness, deep learning functionalities, customization, and optimization.

In this course, you will cover the fundamentals of classification and object detection models and apply them to actual datasets using PyTorch. You’ll learn popular architectures and how to implement and fine-tune them for better results. Finally, you’ll learn to convert models to ONNX and OpenVINO to deploy in edge devices.

By the end of this course, you will have acquired the necessary skills to be able to use PyTorch for image identification and object detection in real-world applications

Using PyTorch for Image Classification and Object Detection

If a prediction is correct with its bounding box for class $x$, it is a true positive for class $x$ in object detection.

If a prediction is missed for an object belonging to class $x$, it is a false negative for class $x$.

If the model detects an object as class $x$ but it belongs to class $y$, that prediction is a false negative for class $x$ and a false positive for class $y$.

If the model predicts an object belonging to class $x$ as class $x$ and not as $y$, it is true positive for class $x$ and true negative for class $y$.

**(Select all that apply.)** Which statement is _true_?

(Select all that apply.) Which statement is true?

What is the expected FPS for a model to be real-time?

What is the expected FPS for a model to be real-time?

It is the box predicted by the model for an object.

It is the region proposed by the model before going to the classification head.

It refers to the true answer for an object, including the class label and the object box in object detection, and only the class label in image classification.

It refers to the proportions of the true answers that the model predicted for a dataset.

What does ground truth mean?

It is used to eliminate the bounding boxes encountering the same object.

Between two boxes, if the intersection over union (IOU) is smaller than the given threshold, then one box should be eliminated.

Between two boxes, if the intersection over union (IOU) is bigger than the given threshold, then one box should be eliminated.

It is a postprocess method needed in object detection, not in image classification.

**(Select all that apply.)** Which statement is _true_ about non-maximum suppression?

(Select all that apply.) Which statement is true about non-maximum suppression?

Non-maximum suppression refers to eliminating bounding boxes according to the IOU and preventing multiple bounding boxes from encountering the same object.

An anchor box is the prediction of the model.

Between the R-CNN, Fast R-CNN, and Faster R-CNN architectures, only Faster R-CNN is in the real-time object detector category.

ROI pooling is different than maximum and average pooling. The output from ROI pooling is the same size as the pooling kernel.

Test your understanding of two-stage detection architectures.

Before We Start

Basics of Convolutional Neural Networks

Popular Neural Network Architectures for Image Classification

Using PyTorch for Image Classification

Model Deployment

Basics of Object Detection

Two-Stage Object Detection Architectures

One-Stage Object Detection Architectures

YOLOv7 Model Train and Inference on Edge

Conclusion

Appendix

Quiz on Two-Stage Object Detector