Make your winter worthwhile by joining this exciting Winter Camp project! In this project, you’ll have the opportunity to reconstruct 3D models of real-world objects from video data. You'll learn how to process video frames, generate OBJ models, and refine them using Blender to create high-quality, realistic renderings—all while diving deep into computer vision and modeling.
- Learn the fundamentals of Python programming.
- Set up Git for version control and manage your project code efficiently.
- Understand how to process video data using OpenCV.
- Extract video frames and prepare data for 3D modeling.
- Explore video metadata.
- Analyze frames for important features required for generating 3D models.
- Integrate the steps into a complete pipeline.
- Visualize and refine the 3D models generated from the video data.
By the end of this project, you'll have built a pipeline that:
- Extracts video frames.
- Analyzes metadata.
- Generates 3D models.
- 5-6 hours per week to stay on track with the project tasks.
If you have any queries or need guidance during the project, feel free to contact one of the project mentors listed below:
- Pushpender: 9257009192
- Vikash Sunda: 6376327908
- Mohit Meena: 8448524819
- Prachi Sharma: 6367555641
- Shelly Singhal: 9555340410
We are here to help you, so don’t hesitate to reach out if you need any assistance!
- Lecture one on pandas(see in week one folder)
- Learning Resources:
- Video Tutorial: Python for Beginners (YouTube)
- NumPy Beginner's Guide: Absolute Beginners Guide to NumPy
- Pandas Quick Start: 10 Minutes to Pandas
- Git Introduction: Git and GitHub for Beginners (YouTube)
- Step-by-Step Guide for Git: Access the detailed guide here
Good luck, and let's make this winter challenge a memorable one!