How can I insert my yolo dataset as a table

2 次查看(过去 30 天)
I got a dataset for human tracking, since the pretrained models I found on MATLAB website were for detecting vehicles, but I am struggling to find the way to input my data as ground truth for my model. The dataset I got is ready and cropped. I need to input it as a table. I am using this code and getting the error 'The RowNames property must contain one name for each row in the table.'. I want the table to have 2 columns, one with image path and the other with its data matrix.
myFolder = 'C:\Users\Zewail\Downloads\rois\jpg_Extracted_PIDS';
dinfo = dir(fullfile(myFolder, '**/*.jpg'));
n = numel(dinfo);
T = table(cell(1, n), 'VariableNames', {'image'}, 'RowNames', {dinfo.name});
for K = 1 : n
filename = dinfo(K).name;
filecontent = imread(filename);
T{filename,1} = filecontent;
end

回答(1 个)

yanqi liu
yanqi liu 2022-1-13
yes,sir,please check the demo:
make the dataset as table,and modify model parameter to train
  4 个评论
Mohamed Zewail
Mohamed Zewail 2022-1-13
The dataset I have is already cropped to the ROI, does not need labelling I believe. I am trying to create a table with the same format as vehicleDatasetGroundTruth.mat that is what I need help on. So basically I need a 2 column table, the first column containing the image's path and the second column containg the image as a matrix. This is the code I am using but I am getting the error "The RowNames property must contain one name for each row in the table."
myFolder = 'C:\Users\Zewail\Downloads\rois\jpg_Extracted_PIDS';
dinfo = dir(fullfile(myFolder, '**/*.jpg'));
n = numel(dinfo);
T = table(cell(1, n), 'VariableNames', {'image'}, 'RowNames', {dinfo.name});
for K = 1 : n
filename = dinfo(K).name;
filecontent = imread(filename);
T{filename,1} = filecontent;
end

请先登录,再进行评论。

类别

Help CenterFile Exchange 中查找有关 Sequence and Numeric Feature Data Workflows 的更多信息

产品


版本

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by