[TUTORIAL] Pedestrian Detection

This is a tutorial article on using the Pedestrian Detection application in Deep Block.

In this tutorial, we will detect pedestrians in the images.

Let’s try!

  1. Sign in, and click “TRY DEEP BLOCK” button on the top right corner.
  2. At the Console page, click “Create” button on the top right corner.
  3. Click Computer Vision and then Click Pedestrian Detection. Set the Project Name as “Penn_Fudan_Pedestrian_Detection”. And then click “Create”.
  4. Open the Project in the Console page.
  5. Drag and Drop the images you want to predict into the “Predict” pane
  6. You’ll see the uploaded images in the Predict pane.
  7. Now you can click the “Predict” button on the top right to start the inference. We already trained the model for pedestrian detection, so you don’t need to train the model for inference.
  8. If you upload a lot of images, it may take a while. If you see the loading sign like below, it means it is still running. You can close the browser while it runs.
  9. You will get a notification when the predicting is finished.
  10. You can see the result by clicking the image files in the Predict pane. You can see that the pedestrians are marked with the bounding box. The detection performance is generally pretty good.
  11. We also provide API service, which allows you to integrate human-detecting AI into your service with just a few lines of code. Feel free to contact us for more details.


Leave a Comment