Sitemap

A list of all the posts and pages found on the site. For you robots out there, there is an XML version available for digesting as well.

Pages

Posts

COLMAP in practice

13 minute read

Published:

Two hundred tourists photograph a statue, and from the photographs alone you can work out where each of them stood. That is Structure-from-Motion, and COLMAP is the program that does it — explained from scratch, then in practice.

実務としての COLMAP

1 minute read

Published:

二百人の観光客が彫像を撮る。その写真だけから、誰がどこに立っていたかを言い当てられる。それが Structure-from-Motion であり、COLMAP はそれを行うプログラムです。ゼロから説明し、そのあと実務の話へ。

বাস্তবে COLMAP

11 minute read

Published:

দুইশো পর্যটক একটি ভাস্কর্যের ছবি তোলে, আর কেবল সেই ছবিগুলো দেখেই বলে দেওয়া যায় কে কোথায় দাঁড়িয়ে ছিল। এটাই Structure-from-Motion, আর COLMAP হলো সেই প্রোগ্রাম যা কাজটি করে — একেবারে গোড়া থেকে, তারপর হাতেকলমে।

What is a Gaussian Splat?

5 minute read

Published:

The usual way to turn a point cloud into a model is to stretch a surface over it. Plants punish that assumption — so here is the alternative.

Gaussian Splat とは何か

less than 1 minute read

Published:

点群をモデルにする定石は、その上に面を張ることです。しかし植物は、その前提を容赦なく突き崩します。ではどうするか。

Gaussian Splat কী?

5 minute read

Published:

বিন্দুমেঘ থেকে মডেল বানানোর চেনা পথ হলো তার উপর একটি তল টেনে দেওয়া। গাছ সেই ধারণাটিকেই ভেঙে দেয় — তাই বিকল্পটি কী, সেটাই এখানে।

How a computer sees a plant in 3D

4 minute read

Published:

Walk around a plant with a camera and you can rebuild it in three dimensions. Here is how that actually works — and what it quietly fails to tell you.

What is plant phenotyping?

4 minute read

Published:

The difference between a plant’s genes and how it actually grows — and why measuring the second one is surprisingly hard.

植物フェノタイピングとは何か

less than 1 minute read

Published:

植物の遺伝子と、実際の育ち方との違い。そして後者を測ることが、なぜこれほど難しいのか。

portfolio

3DGS Pipeline Manual Permalink

Bilingual (English / 日本語) end-to-end manual for the 3D Gaussian Splatting capture-to-phenotyping pipeline: environment setup, video processing, COLMAP SfM, 3DGS training, rendering, and automated trait extraction — with parameters and troubleshooting.
Open the manual →

publications

A Low-Cost Autonomous UGV with Self-Charging Capability for Efficient Greenhouse Monitoring

Published in IEEE GCCE 2025, Osaka, Japan, 2025

A low-cost, self-charging autonomous ground vehicle for greenhouse monitoring, achieving 100% autonomous docking and direct-contact charging 58.4% faster than wireless.

Recommended citation: AL Zobaer, et al. (2025). "A Low-Cost Autonomous UGV with Self-Charging Capability for Efficient Greenhouse Monitoring." IEEE GCCE 2025, Osaka, Japan.

Scale-Invariant Render-Space Phenotyping for Time-Series Crop Monitoring with 3D Gaussian Splatting

Published in Computers and Electronics in Agriculture (under review), 2026

A scale-invariant, render-space method that measures plant traits from 3DGS renders, achieving a 2.86× improvement in cross-session stability (CV 28.0% → 9.8%), validated over 49 days in a commercial greenhouse.

Recommended citation: AL Zobaer, et al. (2026). "Scale-Invariant Render-Space Phenotyping for Time-Series Crop Monitoring with 3D Gaussian Splatting." Computers and Electronics in Agriculture (under review).

talks

teaching