YOLO-NAS The Ultimate Course for Object Detection & Tracking

Explore how to perform inference with YOLO NAS on images, videos, and webcam feeds, test the Uranus model on random images and videos, and export NAS models in onnx format.

Master object detection on images with yolo nas small and coco weights in a PyCharm setup, using OpenCV and SuperGradients to run on CPU and produce labeled outputs.

this lecture shows object detection on videos using yolo-nas, with python and opencv code to run on input videos (e.g., bikes.mp4) and save annotated outputs to output.mp4.

learn to perform object detection on a live webcam feed with YOLO-NAS, compare its small, medium, and large models, and display bounding boxes with labels using a 0.7 confidence threshold.

Run YOLO-NAS large in Google Colab to perform fast object detection on images and videos using GPU, with model architecture summary and COCO pretrained weights.

Train YOLO-NAS for pothole detection using a dedicated dataset. Annotate with Roboflow, fine-tune Uranus in Google Colab, and test on images and a demo video.

Train a Yolo-nas small model on a personal protective equipment dataset in Google Colab, exporting data from Roboflow, and evaluate with mean average precision at IOU 0.5.

Explore real-time object detection with YOLO NAS on a live webcam, testing pre-trained coco weights and a custom PPE model, and export both to the ONNX format.

Explore object detection and tracking using the Uranus and SORT algorithms, starting from reading images and videos to detecting objects and tracking them, then counting vehicles entering and leaving.

Learn to read and display an image with OpenCV in Python, set up a PyCharm project, install the OpenCV Python package, and resize the image for full-screen viewing.

Learn to read and display video using OpenCV Python, including setting up a video directory, using cv2.VideoCapture, reading frames, showing them with cv2.imshow, and resizing for display.

Learn to read, display, and save video using OpenCV in Python, using cv2.VideoWriter to output files and handling frame size, frames per second, and basic live webcam capture.

Learn how to capture video from a camera using OpenCV in Python, save the captured footage, and set up the webcam input before exploring object detection with YOLO-NAS.

Learn to perform object detection on images using YOLO-NAS with the super gradients package, running a small model on CPU and visualizing results with OpenCV.

Learn to detect objects in videos with YOLO-NAS using OpenCV and the super gradients model, draw bounding boxes with confidence scores and coco-class labels, and prepare for deepsort tracking.

Track detected objects with the SORT algorithm integrated with YOLO-NAS object detection, assigning a unique ID to each object across video frames and the live webcam feed.

Count vehicles entering and leaving with YOLO-NAS for detection and SORT for tracking, using line crossing to update total counts and display results.

Train and fine-tune YOLO-NAS on a custom ship dataset, then track ships in satellite images with the sort algorithm, covering dataset preparation, augmentation, and evaluation.

Build a Streamlit-based SEO article writer powered by ChatGPT, integrating the OpenAI API to generate keyword-driven articles (e.g., object detection) with configurable word counts and downloadable txt files.

Train a YOLO-NAS model on a 26-class vegetable dataset with augmentation to balance classes and detect vegetables in images and videos for a streamlit app using ChatGPT to generate recipes.

Create a streamlit app that runs YOLO-NAS vegetable detection on video feeds and uses ChatGPT to generate recipes from detected vegetables, featuring an about page and a recipe generator.

Discover the segment anything model from Meta, an image segmentation tool that identifies and extracts objects in images and videos with zero-shot performance.

Combine yolo-nas object detection with the segment anything model to generate masks, focusing on the person class, using coco dataset labels and bounding boxes for precise segmentation.

Fine-tune YOLO-NAS on the American Sign Language letters dataset and test on images, videos, and the live webcam. Export the trained model to ONNX format.

Fine-tune a yolo-nas small model on a three-class face mask dataset (mask, without mask, mask worn incorrectly) to enable real-time detection and video evaluation.

Fine-tune the YOLO-NAS small model on a fire detection dataset with 1322 images, then test best weights on demo videos to detect fire with bounding boxes.

Detect and count vehicles in a defined area using YOLO-NAS for detection and SORT for tracking. Use a mask and line crossing for counting.

Implement a robust people counter by combining YOLO-NAS object detection with SORT tracking to count entering and leaving people in a defined area.

Learn to perform privacy blurring of detected objects with YOLO-NAS, set up a PyCharm project, install OpenCV Python and super gradient, run object detection on video, and blur bounding boxes.

Learn license plate detection with bounding boxes using YOLO NAS small, fine-tune on a single-class dataset, then read plate text with easy OCR in images and videos.

Learn to detect license plates with YOLO-NAS and read them with EasyOCR in an end-to-end ANPR workflow, including project setup and displaying OCR results.

Learn to count plastic bottles on manufacturing lines by detecting with yolo-nas, tracking with sort, and counting when objects cross a line in this end-to-end tutorial.