Original subtitles

As previously explained, there are brand new features that were not available in the previous version

of YOLO.

The first one is post estimation.

So in this video, we'll show you how to run process automation with your office seven.

In the official repository code is provided to run post estimation.

The code is written in the following Python notebook.

However, there are three disadvantages.

First, you have to install Jupyter notebook on your laptop or PC.

Second, the speed is slow on an 810 by 1080 image.

Post estimation takes 3 to 6 seconds, so you cannot use it to do post estimation on video.

Therefore, in this lecture we will not use the notebook.

Instead of using this notebook, we will sell you a better way in doing post estimation using your office.

Seven.

In the official repository.

The code for post estimation is not in the main brands but in post brands.

To enter the brands.

Click the following dropdown, then click posts.

Here's the python code for post estimation.

In the repository that we use.

We have provided Python code to make it easier for you to do post estimation.

The code can be found in the text.

The PI.

The code in the detectors.

The PI file is taken from the text pi in the post brand's.

However, there are some adjustments in the tech post the pie.

The first step in doing post estimation is to download yolo v seven weights.

Continue to scroll down.

Here underweight specifically yolo 57w6 posts.

Download weights by clicking the following link like this.

Wait until the download is finished.

When you're finished.

Navigate to the downloads folder.

Next move is to the YOLO of seven CPU Salute folder.

Click the file, then press control X.

Pace in the fall of 73 use route followed by pressing control fee.

We have provided pictures and video that can be used for post estimation.

The inmates and video can be found at the following you are.

Next download the image.

Then also download the video.

Wait until the download is finished.

When finished.

Navigate to the downloads folder.

Next move the following image and video to the inference folder in the URL of seven CPU's root folder.

Look like this then press control x.

Place in the conference, followed by pressing control fee.

The next step open any kind of prompt.

Press the windows key, then type in a condom.

Click on any kind of prompt after that.

Activate the of seven CPU environment.

Use the command.

Activate yolo v 7 to 4 you and V.

Press enter.

Then navigate to the YOLO fee seven zip use route folder.

First, we will run cost estimation on the image.

Calls the P.A. is the image that will be used.

Use the following command to run post estimation.

Python detect post the p.

That's that's why it's yolo v7w6 posts.

That's the source.

Inference Postdoc P and G.

That's the escaped label.

That's does feel imt.

Press enter.

The first estimation results will be saved in the folder below.

We open Windows Explorer to see.

Return to the goal of his 74 year old father.

Then navigate to the folder containing the detection results.

The pulse estimation result is shown below.

There is a confidence score bounding box and key point.

Next is the first estimation on the video.

Cos video dump four is the video that could be used.

Use the command below.

Python detective was the key.

That's that's why it's yolo v7w6 posts.

That's that source inference post video dot MP for.

Thus does kept labor.

Das does few imt.

Press internal.

Here are the results of the post estimation on the video.

The post estimation results are very good for the three person objects in this video.

Congratulations.

You can run post estimation using your lucky seven.

See you in the next video.