YOLO Introduction
YOLO (You Only Look Once) is an advanced real-time object detection algorithm primarily used for visual recognition. It was proposed by Joseph Redmon et al. in 2015. Its core idea is to treat the object detection task as a single regression problem, directly predicting the bounding box coordinates and category of objects through a single forward pass of the network, significantly improving detection speed and efficiency.
YOLO11 Classification
This model classifies images, outputting the probability of what an object is.
YOLO11 Detection
This model finds all annotated object types in an image.
YOLO11 Oriented Detection
This model finds all annotated object types in an image. In addition to the detection model's capabilities, it also detects the rotation angle of the object detection box.
YOLO11 Image Segmentation
This model builds upon the detection model by adding the ability to determine which pixels belong to the target object.
YOLO11 Pose Estimation
This model identifies poses, outputting coordinates of keypoints such as nose, eyes, ears, shoulders, hands, and legs. These coordinates can be used to determine postures. You can also train it to recognize poses of other animals.
Model Conversion
This section explains how the .nb format model files used in the previous YOLO11 examples running on WalnutPi are converted. This tutorial requires some basic Linux system knowledge.
Online Model Training
The WalnutPi online model training website is designed for users who need to train their own visual models. Without any coding, simply drag and drop images to build a dataset, and with a few simple steps, you can generate a powerful YOLO11 visual model. Upon completion of training, model files and Python scripts are automatically generated and can be directly run on the development board.