jevIMAGE

3DGS spatial classification prototype uses Jev to judge indoor/outdoor and passability in under a second

A prototype renders 3DGS with PlayCanvas, runs segmentation and object detection in the browser via LiteRT.js, and uses Jev to decide indoor/outdoor, vegetation presence, and passability in under one second.

ダックビル＠STUDIO DUCKBILL LLC

@DuckbillStudioArchived 2026.09.21Originally posted 2026.09.20

3DGS spatial classification prototype uses Jev to judge indoor/outdoor and passability in under a second (image 1 of 2)

3DGS spatial classification prototype uses Jev to judge indoor/outdoor and passability in under a second (image 2 of 2)

ダックビル＠STUDIO DUCKBILL LLC @DuckbillStudioOriginally posted 2026.09.20

Open original on X

3DGS空間の「屋内/屋外」「植生の有無」「通行可否」などを、1秒未満で判定するプロトタイプを作りました。 TypeSafe AIの「Jev」を使った高速な空間分類です。 PlayCanvas（3DGS描画） ⬇️ LiteRT.js（ブラウザ内でセグメンテーション＋物体検出） ⬇️ Jev（判定） ⬇️ 結果表示

x.com/duckbillstudio/status/2101582597591834742Archived 2026.09.21 10:28