TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Shiv Integer (2016)

1 点作者 matthberg将近 3 年前

1 comment

dekhn将近 3 年前
Is anybody aware of any recent machine learning work in 3d object composition space? In particular, I often waste a lot of time manually finding parameters that work for a particular system and it seems like if I just had a differentiable system for working with things like 3d primitives (spheres, cylinders, etc) I'd be able to explore that space with constraints efficiently.