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。

Show HN: Moonshine – open-source, pretrained ML models for satellite

86 点作者 nharada大约 2 年前
Hey HackerNews,<p>Today I&#x27;d like to share my open source project, Moonshine!<p>Pretrained vision models are a popular way to reduce how much data you need and to speed up training, but for remote sensing (i.e. aerial, satellite) it can be a challenge to find good weights. Why use Moonshine?<p>1. Pretrained on multispectral data: Existing popular CV packages are nice to use but are trained on ImageNet or similar, meaning that not only is the domain of the pretraining different, but you may be restricted to only 3 channels. For remote sensing platforms with multispectral data, this might be a non-starter. Moonshine includes models specifically trained for your remote sensing problems.<p>2. Focus on usability: There are academic releases of specialized models that do support remote sensing data, but often they are difficult to use. They might be hard to install, and often are supported by a grad student who is more focused on publication than software. A core tenant of Moonshine is that it should be easy to use.<p>I&#x27;ve been working on this project for nearly a year now and I&#x27;m really excited to show it off and most importantly get feedback! I have big plans for what to build in the future, but this set of features was the smallest one I could think of that would provide some value.<p>Docs: <a href="https:&#x2F;&#x2F;moonshineai.readthedocs.io&#x2F;en&#x2F;latest&#x2F;index.html" rel="nofollow">https:&#x2F;&#x2F;moonshineai.readthedocs.io&#x2F;en&#x2F;latest&#x2F;index.html</a> Github: <a href="https:&#x2F;&#x2F;github.com&#x2F;moonshinelabs-ai&#x2F;moonshine">https:&#x2F;&#x2F;github.com&#x2F;moonshinelabs-ai&#x2F;moonshine</a>

8 条评论

ravery大约 2 年前
It&#x27;d be great to have a page comparing this to other geoML libs. TorchGeo provides pretrained model weights for Sentinel-2: <a href="https:&#x2F;&#x2F;torchgeo.readthedocs.io&#x2F;en&#x2F;stable&#x2F;tutorials&#x2F;pretrained_weights.html" rel="nofollow">https:&#x2F;&#x2F;torchgeo.readthedocs.io&#x2F;en&#x2F;stable&#x2F;tutorials&#x2F;pretrain...</a><p>there&#x27;s an existing comparison here: <a href="https:&#x2F;&#x2F;github.com&#x2F;weiji14&#x2F;zen3geo&#x2F;discussions&#x2F;70">https:&#x2F;&#x2F;github.com&#x2F;weiji14&#x2F;zen3geo&#x2F;discussions&#x2F;70</a>
评论 #35270527 未加载
tectonic大约 2 年前
You should definitely submit this to <a href="https:&#x2F;&#x2F;github.com&#x2F;orbitalindex&#x2F;awesome-space#earth">https:&#x2F;&#x2F;github.com&#x2F;orbitalindex&#x2F;awesome-space#earth</a>!
评论 #35267345 未加载
yeldarb大约 2 年前
This is really neat. Is the idea that people will finetune from these pre-trained weights?<p>Would be really curious to learn the delta in model performance on downstream tasks relative to training from randomly initialized weights or generic weights like COCO.<p>We released a benchmark dataset of real-world tasks shared by users recently that has several aerial detection datasets you might be able to use to measure this: <a href="https:&#x2F;&#x2F;www.rf100.org" rel="nofollow">https:&#x2F;&#x2F;www.rf100.org</a><p>(And if you need help this sounds like exactly the type of thing we’d be interested in collaborating on!)<p>Edit: we also have a bunch more aerial datasets listed here that could be useful for this: <a href="https:&#x2F;&#x2F;universe.roboflow.com&#x2F;browse&#x2F;aerial">https:&#x2F;&#x2F;universe.roboflow.com&#x2F;browse&#x2F;aerial</a>
rapiz大约 2 年前
Can someone explain what are these models for? For a person who knows nothing about geo data, it looks like something that helps to process the data that satellites capture.
评论 #35272283 未加载
评论 #35271988 未加载
dalekchef大约 2 年前
In general is it okay to use datasets that have NON-Commercial clauses in them to promote a commercial business? I guess because you are not selling the models it is okay?<p>I am not a lawyer, just curious from an ML research perspective.<p>From: <a href="https:&#x2F;&#x2F;github.com&#x2F;fMoW&#x2F;dataset&#x2F;blob&#x2F;master&#x2F;LICENSE">https:&#x2F;&#x2F;github.com&#x2F;fMoW&#x2F;dataset&#x2F;blob&#x2F;master&#x2F;LICENSE</a><p>NonCommercial means not primarily intended for or directed towards commercial advantage or monetary compensation. For purposes of this Public License, the exchange of the Licensed Material for other material subject to Copyright and Similar Rights by digital file-sharing or similar means is NonCommercial provided there is no payment of monetary compensation in connection with the exchange.
评论 #35270520 未加载
johnea大约 2 年前
Actully, it&#x27;s not for satellites, it&#x27;s for satellite data. I&#x27;m at the JPL Flight S&#x2F;W workshop in Pasadena right now. There are a lot of talks about ML actually IN satellites. But this S&#x2F;W is strictly ground side... I&#x27;m sure it&#x27;s a valuable tool for data analysis, but a rewording of the topic would be more accurate...
评论 #35269114 未加载
zecken大约 2 年前
Wow this seems really useful! Is it bad the first thing I think when I see this is how a nefarious government can use it with bad intentions?
mistrial9大约 2 年前
How can this get to World Wildlife Fund and other non-specialists doing environmental work ? add new publicity (check!) .. other ways?
评论 #35262790 未加载