YOLOv4 in the CLOUD: Build Object Tracking Using DeepSORT in Google Colab (FREE GPU)
Learn how to build and run your very own Object Tracker in Google Colab! This tutorial walks you through the process of building an object tracking application using DeepSORT and YOLOv4 Object Detection in the Cloud on Google Colab's FREE GPU! Get it up and running with only a few clicks of a button. This tutorial covers it all.
#yolov4 #deepsort #cloud
THE GOOGLE COLAB NOTEBOOK: https://colab.research.google.com/dri...
YOLOv4 is a state of the art algorithm that uses deep convolutional neural networks to perform object detections. We can take the output of YOLOv4 feed these object detections into Deep SORT (Simple Online and Realtime Tracking with a Deep Association Metric) in order to create a highly accurate object tracker.
GET THE CODE HERE: https://github.com/theAIGuysCode/yolo...
In this video I cover:
1. Setting up the Colab Notebook and Enabling GPU
2. Cloning the code and installing dependencies.
3. Converting YOLOv4 pre-trained model into TensorFlow model.
4. Running Object Tracker on video.
5. Filtering allowed classes to track.
6. Adding info flag to see detailed information on tracked objects.
------------------------------Resources------------------------------
Run DeepSORT Object Tracker on Local Machine: • Object Tracking Using YOLOv4, Deep SO...
Learn to Convert to TFLite and TensorRT: • YOLOv4 Object Detection with TensorFl...
Configure to Run with Custom YOLOv4 Detector: • How to Build a Custom YOLOv4 Object D...
Train Custom YOLOv4 Detector in Cloud: • YOLOv4 in the CLOUD: Build and Train ...
The Official YOLOv4 paper: https://arxiv.org/abs/2004.10934
If you enjoyed the video, toss it a like! 👍
To Subscribe: / @theaiguy
Thanks so much for watching!
The AI Guy
Watch video YOLOv4 in the CLOUD: Build Object Tracking Using DeepSORT in Google Colab (FREE GPU) online without registration, duration hours minute second in high quality. This video was added by user The AI Guy 29 September 2020, don't forget to share it with your friends and acquaintances, it has been viewed on our site 24,846 once and liked it 696 people.