I'm also interested in detecting smaller objects like bean. According to the extent of research you have conducted, what model or algorithm is most suitable? Or is there a special layer that you use?
Thanks for your videos. i have question if you can provide how we can you use it for person reidentifaction Tasks. and about your video about Yolo-world can we use it as reidentificator object?
I have a yolov8 course for exactly that. Yolov9 can just be swapped out directly. We go over re-identification and a bunch of trackers and methods for that. Even the theory behind than and trackers in general
This guy works for Ultralytics. Of course he would prefer yolov8 over anything. Take your decision cautiously. I also trained yolov8 model. And when asked for enterprise license, they asked for 5000 USD/annum even for a single person venture with no income. Just posting so people can take decision based on their budge and needs.
I don’t work for ultralytics. I’m just making videos. What’s on my channel has nothing to do with them. You can see the comparisons side by side and I never said yolov8 is better. Actually I’m saying that yolov9 is better but yolov8 is still a lot faster. Yolov9 is based on ultralytics and yolov5, so you won’t be able to use yolov9 either without a license. Have a great day and do your own diligence
@@NicolaiAI Hey Nicolai, your name is on the Ultralytics website on the team page and you also put out videos on the Ultralytics RUclips channel, what do you mean you don't work for Ultralytics ? 😂
Join My AI Career Program
👉 www.nicolai-nielsen.com/aicareer
Enroll in My School and Technical Courses
👉 www.nicos-school.com
Great video! I have been watching your content from quite a few days and all the videos are very informative. Thanks! :)
thanks a ton man! Welcome to the channel
Can you compare YOLOv9 with YOLONas?
I'm personally struggling a lot with small object detection, so it'd be nice to see this comparison
I'm also interested in detecting smaller objects like bean. According to the extent of research you have conducted, what model or algorithm is most suitable? Or is there a special layer that you use?
Thank you for the video!!! what dataset are you using?
These are just random videos from the internet so the models have not seen any of these examples before
Threshold seems to be 0.1 for Yolo V9 and 0.2 fot Yolo V8 on your tests, that 's why Yolo V9 find some false boat
Thanks for your videos. i have question if you can provide how we can you use it for person reidentifaction Tasks. and about your video about Yolo-world can we use it as reidentificator object?
I have a yolov8 course for exactly that. Yolov9 can just be swapped out directly. We go over re-identification and a bunch of trackers and methods for that. Even the theory behind than and trackers in general
what's the performance difference?
Yokov9 slightly better on benchmarks but could be very different depending on specific dataset and use case
Guess v8 still the better at overall
For sure the easier and better framework plus speed
This guy works for Ultralytics. Of course he would prefer yolov8 over anything. Take your decision cautiously. I also trained yolov8 model. And when asked for enterprise license, they asked for 5000 USD/annum even for a single person venture with no income. Just posting so people can take decision based on their budge and needs.
I don’t work for ultralytics. I’m just making videos. What’s on my channel has nothing to do with them. You can see the comparisons side by side and I never said yolov8 is better. Actually I’m saying that yolov9 is better but yolov8 is still a lot faster. Yolov9 is based on ultralytics and yolov5, so you won’t be able to use yolov9 either without a license. Have a great day and do your own diligence
@@NicolaiAI When do i need enterprise license ?
@@NicolaiAI Hey Nicolai, your name is on the Ultralytics website on the team page and you also put out videos on the Ultralytics RUclips channel, what do you mean you don't work for Ultralytics ? 😂