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。

StyleNeRF: A Style-Based 3D-Aware Generator for High-Resolution Image Synthesis

109 点作者 lnyan大约 3 年前

9 条评论

fxtentacle大约 3 年前
Cool idea :)<p>1. They generate a low-resolution position map using NeRF =&gt; Good camera and 3D control<p>2. They then upsample that with StyleGAN2 =&gt; Good features in the final 2D image.<p>So in effect, this is a StyleGAN2 network where the latent space has the 3D-related components split off into a different generator network.
harry-wood大约 3 年前
The twinkling children&#x27;s TV style background music in that video is not making it any less creepy.
motohagiography大约 3 年前
While watching it, I wondered whether the &quot;this x does not exist,&quot; GAN sites also implies a subset of &quot;this x <i>must</i> exist&quot; and if looking at it that way could yield discoveries in other areas. e.g. This X must exist, where x was a solar system, protein, wave function, chemical substance, cell, transcendental number, etc. Naive view, but when you&#x27;re extapolating from a field of arbitrary 2D data, it seems there are a lot of ladders you could climb with that.
egnehots大约 3 年前
that&#x27;s weird that they mostly ignore the comparison to stylegan3: <a href="https:&#x2F;&#x2F;nvlabs.github.io&#x2F;stylegan3&#x2F;" rel="nofollow">https:&#x2F;&#x2F;nvlabs.github.io&#x2F;stylegan3&#x2F;</a><p>it&#x27;s not a nerf based solution but since the goal is high resolution image syntesis..
Allstar大约 3 年前
The fact that this was in part developed with the help of Facebook AI Research is kind of unsettling.<p>That said, the results are actually very impressive.
评论 #30643380 未加载
评论 #30641121 未加载
评论 #30640162 未加载
axg11大约 3 年前
I see there are a lot of questions&#x2F;doubts in the comments about the usefulness of such research. This type of research is the precursor to gaining the ability to generative photorealistic 3D worlds. What is that useful for? Anywhere we currently generate 3D environments, including movies, games, VR&#x2F;AR.
kingcharles大约 3 年前
GANs are getting creepier by the day.
评论 #30637566 未加载
birdyrooster大约 3 年前
Interesting that the reflections in each eye vary so much between each other.
评论 #30638450 未加载
petesergeant大约 3 年前
subtitle: an amazing trip through the uncanny valley