Task:

1. Take photos of your environment of two or more objects. (at least 100 instances between all objects)

2. Annotate them on roboflow.

3. Train a Faster RCNN model using detectron2

4. Train Yolov4/5/6/7/8 (only one of them of choice) the smallest size

5. Evaluate both models based on mAP and speed and size.

1. Taking photos

2 Objects' pictures were taken. 1. Pyramid, 2. Glasses Pyramid image :

Glasses image:

2. Annotation on roboflow

3. A Faster CNN using detectron2

4. Yolov8 trained for 20 epochs

• confusion matrix

• F1 curve

• PR curve

• P curve

• R curve

5. Evaluation and comparison

Mean Average Precision

• Faster RCNN: ~0.01%
• Yolov8: ~0.13%

Speed:

Yolo is more speedier despite its size, and its speed has a number of advantages. Yolo training takes 3 to 4 minutes for 24 epochs, but Faster RCNN (detectron2) takes more than an hour.

Size:

• Bigger RCNN model size: 805.5 Mb
• The Yolov8 model size: 21.53 Mb

References

1. https://docs.roboflow.com

2. https://blog.roboflow.com/how-to-train-yolov8-on-a-custom-dataset/

3. https://blog.roboflow.com/how-to-train-detectron2/

4. https://blog.roboflow.com/how-to-train-detectron2/#using-your-own-data-with-detectron2

LINK TO THE COLAB: Link