Hello, I want to train ssd object detector with own dataset of 1865 images… what do i have to change in the code in this example to run the code corectly? https://www.mathworks.com/help/deeplearning/ug/object-detection-using-ssd-deep-learning.html

Antworten (1)

Sulaymon Eshkabilov
Sulaymon Eshkabilov am 22 Mai 2023

0 Stimmen

If you look at carefully to the documentation, it states explicitly that you'd need to load your data:
...
data = load('vehicleDatasetGroundTruth.mat');
vehicleDataset = data.vehicleDataset;
That means what you should do is to create data store of your images: see DOC how to create data store images.
Step 1. Create a data store from your images using: imageDatastore()
Step 2. Load your created data store (Optional: to create one .mat file and then load the file is optional. Only if you want to keep/store all your created data sets in one *.mat file)
Step 3. Start simulation

1 Kommentar

Hello Sulaymon, yes that’s what i already did. But is there anything else what i need to change? Don’t I have to change the values of anchor boxes somehow? Or can I leave them like it is shown in example?

Melden Sie sich an, um zu kommentieren.

Kategorien

Mehr zu Image Processing and Computer Vision finden Sie in Hilfe-Center und File Exchange

Gefragt:

am 22 Mai 2023

Kommentiert:

am 23 Mai 2023

Community Treasure Hunt

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

Start Hunting!

Translated by