eceo-epfl

eceo-epfl / deepreefmap

Public

Tools for reconstructing and mapping coral reef scenes from underwater images

46
2
100% credibility
Found May 11, 2026 at 46 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

DeepReefMap turns underwater video clips from action cameras into interactive 3D models, overhead images, and statistics of coral reefs.

How It Works

1
📹 Film your reef

Swim over the coral reef with your action camera and record a short video clip of the underwater scene.

2
💻 Set up the tool

Download the free mapping software and get it ready on your computer in a few simple steps.

3
Prepare your camera
✅
Use ready preset

Pick a common action camera like GoPro and go right away.

📐
Calibrate your own

Film a quick calibration clip and let the tool learn your camera's view.

4
🚀 Create the map

Feed in your reef video and watch as the software builds a colorful 3D model live on screen.

5
🔍 Explore in 3D

Click around the interactive viewer to fly through the reef, see labels on corals, and check flat overview and stats.

🌊 Study your reef

Get a detailed 3D map, aerial photo, and health report to understand coral cover and share your findings.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 46 to 46 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is deepreefmap?

DeepReefMap turns underwater videos from handheld cameras like GoPro into semantic 3D maps of coral reefs, producing point clouds, ortho-mosaics, and benthic cover stats in one CLI run. Feed it a clip, pick a camera profile or calibrate your own with COLMAP, choose a mapping backend, and get PLY files for Meshlab, plus an interactive Viser viewer—all in Python with uv for easy deps. It's built for reconstructing reef scenes from images, solving the hassle of manual 3D mapping in marine surveys.

Why is it gaining traction?

The three-command quickstart delivers instant results from raw footage, with resume caching to skip recompute on tweaks, and swappable backends like lightweight SC-SfMLearner or GPU-heavy LoGeR for quality tradeoffs. Interactive Viser lets you inspect frames, toggle classes, and accumulate points live, while outputs like JSON stats make it reef-research ready. Python tools github users love the no-fuss path from video to analysis, beating ad-hoc COLMAP+segmentation pipelines.

Who should use this?

Coral reef ecologists surveying benthic cover with GoPro transects, marine biologists needing 3D models for species mapping, or underwater robotics teams prototyping scene reconstruction. Ideal for field researchers who want ortho products and point clouds without deep CV expertise, especially on Linux or Windows with GPU accel.

Verdict

Grab it if you're in reef mapping—strong docs, CLI, and quickstart make the 1.0% credibility score (46 stars) forgivable for alpha stage. Test on your footage; scale issues may hit long videos, but it's a solid Python github tools foundation for coral deep reef map workflows.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.