Object Detection Model with EfficientNetD0

This is a Python-based object detection model that uses the EfficientNetD0 architecture for object detection. The model can read both images and videos, and generates bounding boxes for detected objects along with the label and percentage of confidence. Some augmentation techniques are implemented into config file of model. For more information -> object_detection&tracking.ipynb

You can access required files and trained models from below links.

Train.record file can be accesed from: https://drive.google.com/file/d/1R4FbsU1B5nJrcBDJxdrfqwJmqJssfbVC/view?usp=share_link

Val.record file can be accesed from: https://drive.google.com/file/d/1Z0Y-G38yJOoEfY9V_Mq1MR0BFV59xY6V/view?usp=share_link

Pre-trained, augmented and fine tuned model can be accesed from: https://drive.google.com/file/d/1yWTv6jVqWMWe9GK-hWqdYKtp_PD6hVL_/view?usp=share_link

Test images(frames) taken from video: https://drive.google.com/drive/folders/1q4sHb0i1JYnaxoFuZfveB3U5GN9xQ8yA?usp=share_link

Train images(frames) taken from video: https://drive.google.com/drive/folders/1UkGSc5CQ-cXsOf4QfQB7uBJy-RuwYtnT?usp=share_link

Validation images taken from video: https://drive.google.com/drive/folders/1wnUdGOPv1Svj9Gxi18LJflqkcC0cTa8u?usp=share_link

Train_video: https://drive.google.com/file/d/1x9oUzGBKj62Ch3Ry2K5TGb8WCb0iK9zF/view?usp=share_link

Val_video: https://drive.google.com/file/d/1tI3sAH0COWS6lz4tSc9tKgoZmxGyznxH/view?usp=share_link

Test_video: https://drive.google.com/file/d/1swHZCEPwbDytkjjAvzfkkUZMM3tS6dkv/view?usp=share_link

Model prediction on test video: https://drive.google.com/file/d/1GYTLOkyiouEkOTBdkTs22nZnsh49ixjm/view?usp=share_link