AK(@_akhaliq)
LEGO-Anything Coding Agents for 3D Scene Reconstruction paper: https://t.co/lWpBbRNuP8
6.4内容质量
TL;DR · AI 摘要
Title: AK @ akhaliq on X URL Source: Published Time: 2026-09-30T22:51:45.000Z Markdown Content: AK on X: "LEGO-Anything...
核心要点
- 主题聚焦：LEGO-Anything Coding Agents for 3D Scene Reconst
- 来源：AK(@_akhaliq)，建议结合原文判断细节。
- AI 分析暂不可用，本条为保底评分与摘要。
#AI#编程
打开原文AK on X: "LEGO-Anything Coding Agents for 3D Scene Reconstruction paper: https://t.co/lWpBbRNuP8" / X
[](https://x.com/)
[](https://x.com/)
AK on X: "LEGO-Anything Coding Agents for 3D Scene Reconstruction paper: https://t.co/lWpBbRNuP8"
-  AK @_akhaliq LEGO-Anything Coding Agents for 3D Scene Reconstruction paper: huggingface.co/papers/2609.36… [Video 2](blob:https://x.com/1799f3ed-21fa-4e18-b5be-b0a0da9bf5f0) 00:50 10:51 PM · Sep 30, 2026·95.6K Views 15 99 670 666
-  Michael Black @Michael_J_Black Oct 1 Why is this called "LEGO"? This is image-to-code, like VIGA. There are several papers on actually producing LEGO constructions of the world. Maybe the earliest, pre-deep-learning, approach is from Anton van den Hengel's group: cs.adelaide.edu.au/~hengel/?/rese… A recent example from Show more [](https://x.com/Michael_J_Black/status/2105513174728397298) 2 17 1.8K
-  Garrett Johnson @garrettkjohnson Oct 1 > says LEGO > looks inside > no LEGO :( 1 1 14 795
-  Jatin Garg @jatingargiitk Oct 1 reconstruction from partial scans is usually the hard part. how does the coding agent handle missing geometry or occlusions? 1 2 412
Log in or sign up for X
See what’s happening and join the conversation
Continue with phoneContinue with Apple
Continue with Google
通过 Google 继续操作 通过 Google 继续操作。在新标签页中打开
or