Shop Traffic Counter On the Web | Computer Vision Web Development
Vložit
- čas přidán 1. 09. 2023
- This is part of the Computer Vision Web Development Course.
bit.ly/3EmdhNB
🚀🚀 My Urdu/Hindi AI CZcams Channel 🚀🚀
/ @murtazahassan01
Premium Courses:
✔️ Computer Vision Web Development:
bit.ly/3EmdhNB
✔️ Computer Vision Game Development Course:
bit.ly/3ttLZ2s
✔️ Computer Vision with Arduino Course:
bit.ly/3wzLB4m
✔️ Advanced Drone Programming Course:
bit.ly/3qs3v5g
✔️ Learn to Build Computer Vision Mobile Apps:
bit.ly/3uioY1J
✔️ Jetson Nano Premium Course:
bit.ly/3L8uIlF
⚙️⚙️⚙️--My Gear - ⚙️⚙️⚙️
👉 Complete Gear 💈: www.computervision.zone/tech-...
👉 My PC Specs 🖥️: www.computervision.zone/tech-...
👉 My Video Shooting Gear📽️ : www.computervision.zone/tech-...
👉 My Laptops 💻: www.computervision.zone/tech-...
👉 Educational Products🧑🎓: www.computervision.zone/tech-...
👉 CZcams Starter Kit 🔴: www.computervision.zone/tech-...
Follow Me:
TikTok: bit.ly/3Vo76OQ
Facebook Group: bit.ly/3irDcb7
Discord: bit.ly/3JvyxAM
Facebook Page: bit.ly/3IvpU7W
Instagram: bit.ly/3NdGME3
Website: bit.ly/3ICFTS0
GitHub: bit.ly/3woU6PS
#ComputerVision
#OpenCV
#CVZone
Why was that guy not detected? and how can we solve the issue?
Comment below here and I will share a heart on the correct answer.
because he's wearing a cap, and the position of this character with his cap on when he arrives at the level of the two reference lines didn't allow him to be recognized continuously until he reached the status ''IN''; because the combination of the character + his cap; on this moment, didn't look like a person according to the trained model, so the algorithm didn't recognize it as a person entering.
Solution:
one solution would be to add an image case among the images when training the model; but this could corrupt the character recognition model generated after training, because this case doesn't really look like a character; and the algorithm could recognize certain objects other than a character later on;
So the best solution, in my opinion, would be to adjust the camera position, so that the algorithm can continuously recognize that it's a character until it passes over the two reference lines.
Maybe because the tracking wasn’t good enough so it lost the person and didn’t understand properly it crossed the line. A solution can be increase the frame of tracking so that if the person is lost for some frame we can still be able to track properly
The bounding box wasn't present the entire time as it crossed the green lines. So your code couldn't assign the object a direction. Maybe you can average the position a bit.
@@jkprogrammeur5525 correct
Great content
Hi Bro
I am from Uzbekistan 🇺🇿
Good luck
The person was not detected because the I'd will assume to every detection and within the particular the same I'd enters again it won't be detection and this is done by sorting algorithm this concept is explained by you already in your 4hrs 101 object detection course 🙂
Although person got detected, but the head of the person was not. You are counting the people only when their head is passing the boundary, way before there entire body or the centre of the bounding box is passing the border. So you must be tracking the heads instead the of bbox center which is good idea, but your head detection dataset doesn't have caps. You just have to add this in your dataset.
Correct. It needs more data with people wearing caps.
The guy was not detected because the trained model on which inference is being done didn’t or had a very small portion of dataset with people with hats.
Or if traditional image processing techniques are being used then simply their were not enough black hair being shown to the camera to detect a head.
Just continued the video and confirmed it was a trained model.
So disregard my earlier comment second part.
Correct. It needs more data with people wearing caps.
is there any chance to port this code to javascript or it's only available in python?
For web, Did you used Django.
We used flask
@@murtazasworkshop no issues I need to lear cloud vision it my aim.
How can I join your class I am from Bangladesh.
Please add me.
@@murtazasworkshop waiting for you kind of response.
Maybe due to cap?
Is this part of Computer Vision Web - Advanced course ?
Its part of the Computer Vision Web Apps . the highest Tier
Thanks
Is the backend part in Django?
The person not detected because he wear hat, we need train the model for such as situation
Employee not customer
Because you did not have people with cap in your training dataset
Correct
Because he was wearing the cap ..not recognised as a person
Amazing as usual. Could you provide the dataset and model architecture used? @murtazasworkshop