{"slug": "what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-world", "title": "What I Learned About 3D Reconstruction in Week 1 of My AI Internship at PreserveMy.World", "summary": "An intern at TechRealm x PreserveMy.World compared five 3D reconstruction methods for preserving cultural heritage sites, including COLMAP, NeRF, Gaussian Splatting, monocular depth estimation, and multi-view stereo. The intern wrote a Python script simulating monocular depth estimation and generated a depth map, confirming the pipeline concept. The nonprofit uses AI to document historical landmarks like Lahore's Badshahi Masjid.", "body_md": "This week marked the beginning of my internship at TechRealm x PreserveMy.World —\n\na nonprofit using AI to preserve cultural heritage sites in 3D. I'm on the Core Tech\n\ntrack, focused on Platform & Web Engineering.\n\nPreserveMy.World is a nonprofit initiative that uses AI and 3D reconstruction to\n\ndocument and preserve historical sites — starting with Lahore's heritage landmarks\n\nlike Badshahi Masjid.\n\nI compared five methods for reconstructing heritage sites in 3D:\n\n**1. COLMAP (Structure-from-Motion)**\n\nTakes multiple photos and produces a 3D point cloud. Great starting point for PMW.\n\n**2. NeRF (Neural Radiance Fields)**\n\nCreates photorealistic 3D scenes from images. Needs a GPU but produces stunning results.\n\n**3. Gaussian Splatting**\n\nFast, high-quality 3D rendering — ideal for web platforms like PreserveMy.World.\n\n**4. Monocular Depth Estimation**\n\nWorks from a single image or video. Most practical for field use with limited equipment.\n\n**5. Multi-View Stereo (MVS)**\n\nDense reconstruction from multiple calibrated images. Good for building facades.\n\nI wrote a Python script simulating monocular depth estimation and generated a depth\n\nmap output. Small experiment, but it confirmed the pipeline concept works.\n\nThe first attempt at downloading a heritage image in Python returned a 403 error —\n\nso I switched to a simulated depth map instead. Real-world data acquisition will need\n\nproper API access or local images.\n\nI'll continue researching 3D reconstruction methods, run more experiments,\n\n*This post is part of my TechRealm x PreserveMy.World internship journey.\nGitHub: https://github.com/Anfey-SE/PMW-day1*", "url": "https://wpnews.pro/news/what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-world", "canonical_source": "https://dev.to/amna_hafeez/what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-preservemyworld-32fd", "published_at": "2026-06-26 19:23:40+00:00", "updated_at": "2026-06-26 20:04:07.480760+00:00", "lang": "en", "topics": ["computer-vision", "artificial-intelligence", "machine-learning", "ai-research", "developer-tools"], "entities": ["TechRealm", "PreserveMy.World", "Badshahi Masjid", "COLMAP", "NeRF", "Gaussian Splatting"], "alternates": {"html": "https://wpnews.pro/news/what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-world", "markdown": "https://wpnews.pro/news/what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-world.md", "text": "https://wpnews.pro/news/what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-world.txt", "jsonld": "https://wpnews.pro/news/what-i-learned-about-3d-reconstruction-in-week-1-of-my-ai-internship-at-world.jsonld"}}