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。

Ask HN: Can an AI model be used to downscale an image?

3 点作者 gampleman超过 1 年前
Things like favicons need a specific resolution. You may have a nice much larger source image (i.e. 4000x3000), but you need it to be easy to see at say 64x64 pixels. You could simply downscale it, but generally speaking the results won&#x27;t be great since to for the image to be readable at such a small resolution, it needs to be simplified. So instead of making tiny blurry lines, fewer, simpler and bolder shapes should be used to represent the same semantic image.<p>Seems like something AI would be good at, but I haven&#x27;t seen anything tackling this problem. Are there any issues to do this? Have I missed some exciting projects tackling this?

3 条评论

gampleman超过 1 年前
To give a better example why I think this is a job for AI. Consider this painting:<p><a href="https:&#x2F;&#x2F;s3.amazonaws.com&#x2F;cdn.monasteryicons.com&#x2F;images&#x2F;popup&#x2F;entry-into-jerusalem-icon-449.jpg" rel="nofollow noreferrer">https:&#x2F;&#x2F;s3.amazonaws.com&#x2F;cdn.monasteryicons.com&#x2F;images&#x2F;popup...</a><p>Now, look at what a cropped and downscaled version of this looks like:<p><a href="https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;Uf66tDO" rel="nofollow noreferrer">https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;Uf66tDO</a><p>You can hardly make out any meaning from the mess of colours in the favicon. A human artist given the job would likely modify the colours of the background to give more emphasis on the central figure. They would also simplify the crowd making it into fewer characters. In general most of the lines would be redrawn to form simpler shapes so that the viewer could actually recognise what the favicon is supposed to represent.
verdverm超过 1 年前
There was an algorithm that does what you want, based on some Terence Tao work iirc, around the time &#x2F; based on his L1&#x2F;L2 norm result...?<p>One of the examples was reconstructing an Obama picture from basic shapes.<p>Struggling to find it again, wisdom of the HN crowd help me out here?
评论 #38713705 未加载
WallyFunk超过 1 年前
If the original image is a vector&#x2F;SVG then it should look well scaled down. I think you can have SVG graphics for a favicon.
评论 #38719318 未加载
评论 #38713709 未加载