Computer Vision Top Projects
Computer Vision has been undergoing rapid advancements in the market, increasingly automating tasks previously done manually. This trend has gained significance in recent years, as numerous technology companies are integrating computer vision into their products, with many already utilizing it.
In 2022, I dove deep into the fascinating realm of computer vision, experimenting with techniques like image classification, object detection, and more. This article is your ticket to my top five hobby projects that not only fueled my passion but also brought in some cash!
YOLOv9 Training On Custom Dataset | Step-by-Step Tutorial for Object Detection using yolov9
Deep Learning Image Classification project using TensorFlow
All right, it’s time to begin our journey 🚀
1- Football Player Detection and Tracking
Before diving into this project, I wasn’t much of a Messi or Ronaldo enthusiast. But one fateful day, while watching a football match, inspiration struck like a lightning bolt. It ignited my curiosity to explore the world of football through the lens of computer vision.
Our journey kicked off with a quest for data, scouring platforms like YouTube and Pixels for match recordings. With unwavering determination, my team and I meticulously annotated over 3,000 images in just one week.
The real excitement came when we initiated model training. To our sheer delight, the initial results exceeded our wildest expectations. But we didn’t stop there — we integrated a tracking module to capture player movements, taking our project’s capabilities to a whole new level.
Fig-1.2: Football Player Detection
Object Detection using YOLOv8:
Image Segmentation using YOLOv8:
2- Fish Detection in Sea
This project’s primary mission? To keep an eye on fish in specific marine spots, all for the benefit of Ichthyologists. The idea struck me after a ’47 Meters Down’ movie night 🎥🌊.
But here’s the kicker: fish data was scarcer than a deep-sea treasure! 🐟💎 Still, I rolled up my sleeves, gathered data from Kaggle, Pexels, and more. Then came the hurdle — annotating! It took me two months! Why? Each fish is unique, like characters in a story. 📚🐠
So, I hit the books and googled my way to understand each fish type.
Fig-1.3: Fish Detection and Tracking
3- Vehicle Detection, Tracking and Counting
This project revolves around the fascinating world of traffic data — detecting, tracking, counting, and analyzing the flow on our roads and streets. Surprisingly, it wasn’t my original idea; rather, an existing concept. What got me excited was its endless applications. I embarked on this as a hobby project, and when I shared it on LinkedIn, I wasn’t sure how it’d be received. To my delight, the response was overwhelmingly positive. That pushed me to enhance it further, adding features like speed calculations and region-based counting.
Fig-1.4: Vehicle Detection, Tracking and Counting
4- Personal Protective Equipment Detection and Monitoring
The heart of my project beats for safety on construction sites! 🏗️ I dove into the world of personal protective equipment to understand the challenges faced by site managers and unravel the mysteries behind workplace injuries.
Fig-1.5: Personal Protective Equipment Detection and Monitoring
5- Virtual GYM Using Pose Estimation
This project was born from a love for home gym workouts. Delving into the realm of fitness tech, I uncovered a surprising gap in the market. While searching for innovative solutions, I stumbled upon YOLO’s pose estimation module — a notable change! 🏋️♀️💡
After immersing myself in weeks of learning and tinkering with this exciting technology, I proudly present a virtual gym experience that’s both fun and functional. Get ready to transform your home workouts with a touch of tech magic! 🏠💪
Fig-1.6: Virtual GYM Using Pose Estimation
My goal was to create a system capable of tracking and quantifying individual exercises within home gym setups. Throughout the project’s development, my primary aim was to address the needs of individuals engaging in home-based fitness routines. When I shared my project on LinkedIn, it garnered interest, resulting in two job offers related to this work.
Thanks 😀😀😀
Conclusion
I constantly explore various ideas and engage in continuous learning across diverse domains where computer vision applications can be beneficial. Executing projects is a hobby of mine, and I relish doing so during my leisure hours, alongside my passion for writing articles.
Thanks for reading; if you liked my content and want to support me, the best way to supporting me on Patreon —
