how load a pre-trained yolov2 net in matlab

Here is an example code to load a pre-trained YOLOv2 network in MATLAB for object detection:


main.m
% Load the pre-trained network
net = load('yolov2ResNet50VehicleExample.mat');
lgraph = net.lgraph;

% Specify the input size of the network
inputSize = [224 224 3];
inputLayer = imageInputLayer(inputSize);

% Replace the input layer of the network with the new input layer
lgraph = replaceLayer(lgraph,'data',inputLayer);

% Set the network training to false
lgraph = setTrainingState(lgraph,'output_1', 'bn_2d_37', false);
lgraph = setTrainingState(lgraph,'output_2', 'bn_2d_44', false);

% Analyze the network layers
analyzeNetwork(lgraph);

% Classify a test image
I = imread('car2.jpg');
I = imresize(I,inputSize(1:2));
[bboxes,scores] = detect(net.detector,I);

% Display the object detection results
if ~isempty(bboxes)
   I = insertObjectAnnotation(I,'rectangle',bboxes,scores);
end  
imshow(I);
804 chars
29 lines

You can replace the inputSize, inputPath, and image name with your own. You can download the pre-trained yolov2ResNet50VehicleExample.mat and use it to load and detect objects in a test image.

similar matlab code snippets

showing picture in every cnn layer in matlab

use mlx which can annotate images as input to do object detection in matlab

create a semantic segmentation algorithm using net = inceptionresnetv2 with imdstest = imagedatastore() and pxdstest = pixellabeldatastore() with image size 224*224*3 in matlab

detect objects in an image in matlab

count number of dots in an image in matlab

autoencoder to remove watermarks in matlab

find corners of object in image in matlab

find a rectangle in an image in matlab in matlab

train a deep network in matlab

train a dlnetwork in matlab

related categories