menu
Real-Time Object Detection & Tracking

Real-Time Object Detection & Tracking

Leader posted 1 min read

During my CodeAlpha internship, I built a system in Python that detects and tracks objects in videos and live webcam streams using YOLOv8 and Deep SORT.

Highlights:

  • Tracks multiple objects with unique IDs
  • Works on video files and webcam input
  • Real-time performance using Python + OpenCV

Challenges I faced:

  • Some objects were “lost” in tracking — required algorithm tweaks
  • Switching between video input types took careful testing
  • Managing dependencies for real-time performance

✨ Seeing it work live was super satisfying!

Check my code and demo here: [GitHub Repo]
https://github.com/Urooj25/CodeAlpha_ObjectDetection.git![](http://)

Python #MachineLearning #YOLOv8 #ComputerVision #DeepSORT #Coding #Internship #Projects

2 Comments

•

This is actually really clean for an internship project.
Did you tweak Deep SORT params or use it mostly out-of-the-box?

1 vote
•

@[CodersMyth] Thanks! I customized Deep SORT parameters like [e.g., max_cosine_distance, max_age, etc.] to improve object tracking accuracy, especially when multiple objects cross paths or move quickly. It really helped make the tracking more stable in both video files and real-time webcam streams.

0
•

Please can you check your github, Thanks!

1 vote
•

@[Dr. Usman Zafar] Thanks for pointing out

0

More Posts

Cavity on X-Ray: A Complete Guide to Detection and Diagnosis

Huifer - Feb 12

5 Web Dev Pitfalls That Are Silently Killing Your Projects (With Real Fixes)

Dharanidharan - Mar 3

Engineering Real-Time Delivery Tracking That Increased Checkout Conversions by 20%

oluwatosinolamilekan - Apr 9

How to save time while debugging

Codeac.io - Dec 11, 2025

Tuesday Coding Tip 05 - Object initialization in C++

Jakub Neruda - Mar 31
chevron_left

More From Urooj Fatima | EE Student

Beyond Hard-Coding: The Engineering Logic of Machine Learning

The Shift from Logic to Learning: An Engineer’s Perspective on ML

Smart Movie Recommender using Linear SVC

Related Jobs

View all jobs →

Commenters (This Week)

7 comments
1 comment
1 comment

Contribute meaningful comments to climb the leaderboard and earn badges!