Taha Yasin Er’s Post

Incredible how the YOLO family of object detection models (yes this is NOT GenAI!) just keeps going 🚀, everytime pushing its capabilities to new levels. Even this new 10th generation still manages to significantly enhance its performance metrics 👏 I clearly remember how incorporating the early version (3ish) in COWI’s national streetview campaign and how it enabled transitioning to fully AI-automated GDPR blurring of faces and licence plates 🥳 Generative AI are without doubt taking all the headlines but the newly coined “Traditional AI” are still the main value driver🦾⚙️

This groundbreaking leap in YOLOv10 performance opens up a realm of possibilities for real-time object detection projects. An improvement that comes mainly from decoupling spatial-channel downsampling! Imagine deploying it for smart city surveillance, enhancing retail analytics, or even creating interactive AR experiences. The potential is limitless, and I can't wait to dive into the world of Computer Vision again. #MachineLearning #ObjectDetection #YOLOv10 #ComputerVision

It's amazing how far the YOLO family has come with each new version. The difference in training models on GPU architectures is clear. Even Apple's architecture can't match a GPU architecture, even using the ‘mps’ device configuration. Adding a single GPU (device=0) already improves the training metrics each epoch the data passes through the model, even if the same learning rate settings, optimization algorithm and dataset are used. On a smaller scale, this proves the superiority of a GPU architecture and may even explain the success of companies such as NVIDIA, which is focused on developing new GPUs.

YOLOv10 just released! 🔥🔥We have a new member and version in the YOLO family. For the nano version of YOLOv10 we are talking 1ms per image (that's 1000 FPS) 🤯 Have not seen such a big jump in performance from any other version release before. Better mAP on the COCO benchmark dataset and close to 2x lower latency compared to the other models. It's based on the Ultralytics framework so we can use it easily in just a few lines of code for both training and inference. We are definitely going to work on some cool videos! Key highlights 🔑 ✅ NMS-free training: Improved performance and reduced latency. ✅ Uses spatial-channel decoupled downsampling for ops efficiency ✅ Adds new compact inverted block (CIB) ✅ Holistic design: Optimized components for efficiency and capability. ✅ YOLOv10: New generation for real-time object detection.

Ultralytics YOLO11 bringing new fashion to life! 🚀😄 What's interesting? ✅ You can track segmented objects, meaning tracking is fully supported. ✅ From my testing, YOLO11 offers everything Ultralytics YOLOv8 does and more, paving the way for advanced computer vision development. Here’s how I built this demo 💪 🚀 `pip install ultralytics` 🚀 `yolo segment predict source='path/to/video/file.mp4'` Excited? Try it yourself! 👇👇👇 ```yolo segment predict source="path/to/video/file.mp4"``` Learn more ➡️ https://lnkd.in/daU8pBPt

Ultralytics YOLO11 - Performance metrics breakdown! 🚀 Let’s talk numbers! YOLO11’s performance metrics speak for themselves: from mAP to processing speeds, it delivers unmatched results. Check out the key stats that make YOLO11 a game-changer in computer vision. Full performance breakdown ➡️ https://ow.ly/p4Kq50TArpj

With meticulous data annotation, I will infuse your machine learning and AI models with unparalleled accuracy and vitality. https://lnkd.in/dAAse5jv #Yankees #Bronny #Hasan #Polymarket #automation #modelling #computervision #aisurgery #data

first time using yolov9 I want to compare it with yolov8 on a complicated project can't wait to see results #objectdetection #computervision #innovation

Boom! Boom! Boom! YOLOv11 is here, and it's packed with impressive upgrades for real-time object detection. They’ve made significant enhancements in terms of: Accuracy: It’s now up to 10% more accurate at detecting objects, delivering even better results. Speed: 20% faster image processing, making it ideal for real-time applications. I can't wait to see how this new version will be used in upcoming projects! Let’s dive into what this release can unlock for us

Chen Junxin and Yu Xiaojie introduced an efficient deep-learning network for road damage detection, the lightweight attention ghost-you only look once (LAG-YOLO). The method makes YOLO more suitable for real-time processing and lightweight deployment while ensuring high accuracy. Explore more：https://lnkd.in/ewbkPyP5