Show HN: AI 照片修复 iOS 应用（设备端处理）
3 分•作者: nicklewers•7 个月前
嗨，各位 HN 用户，
我最近在 App Store 上发布了我的第一个 iOS 应用！它是一个照片清理应用，使用设备端的 Apple Vision 模型将相似的图像聚集成组，然后对每组中的照片进行排名。
链接：[https://apps.apple.com/app/ai-photo-cleaner-sorter/id6754172959](https://apps.apple.com/app/ai-photo-cleaner-sorter/id6754172959)
工作原理：
每组中最好的照片默认被标记为“已保存”，但在“审查”阶段，您可以从该组中选择其他照片进行保存，或者取消选择“最佳”照片并选择另一张保留。总体的理念是“保留最好的，删除其余的”，理论上，如果您同意聚类/评分结果并且不想进行任何修改，您只需轻点几下即可清理数百张照片（和 MB 的存储空间）。
动机：
我就是那种不知怎么就会让相机胶卷变得异常臃肿的人——同一道菜拍了 15 张照片，同一日落拍了 10 个版本。所以我最初开发这个应用是为了帮助自己清理在为期一周的假期中拍摄的 500 张照片……我试用了许多其他应用，但大多数似乎都采用了类似 Tinder 的流程——左滑、右滑，这几乎没有减少工作量。我想要一种有主见的方式来轻松进行大规模清理。
我的路线图上的事项：
* 主要语言的本地化
* 改进的后台处理（这样您就可以清理 1000 多张照片，而无需保持应用打开状态）
* 用户对照片评分计算方式的偏好设置，例如，有些人更看重情感信号而不是图像清晰度。
该应用完全免费使用（也许有一天我会尝试盈利，如果我完全满意的话），我很乐意收到任何反馈——好的和坏的！ （特别感谢您对分组和评分效果的反馈——有任何问题吗？）
谢谢！
Nick
查看原文
Hi HN,<p>I've recently published my first iOS app to the Appstore! Its a photo cleanup app which uses on-device Apple Vision models to cluster similar images together into groups, and then ranks each photo within each group.<p>Link: <a href="https://apps.apple.com/app/ai-photo-cleaner-sorter/id6754172959">https://apps.apple.com/app/ai-photo-cleaner-sorter/id6754172...</a><p>How it works:
The best photo from each group gets marked as "Saved" by default, however in the Review stage, you can select others from the group to save if you want or unselect the "Best" one and choose another one to keep. The general concept is "keep the best, remove the rest", and in theory if you agree with the clustering/scoring and didn't want to make any amendments, you could clean up 100's of photos (and MBs) with just a few taps.<p>Motivation:
I'm one of those people who somehow builds up an insanely bloated camera roll - 15 photos of the same restaurant dish, 10 versions of the same sunset. So I initially built this for myself to help cleanup those 500 photos taken during a 1 week vacation... I've tried a load of the other apps out there, but most seem to have adopted a Tinder style process - swipe left, swipe right, which hardly decreases the workload. I wanted an opinionated way to do a mass cleanup easily.<p>Things on my roadmap:
- Localisation for major languages
- Improved background processing (so you can clean up 1,000+ photos without keeping the app open)
- User preferences for how the photo scoring is calculated i.e. some people value emotional signals over image sharpness.<p>Its totally free to use (maybe one day I'll try and monetise it, if I'm ever fully satisfied )), I'd love any feedback - good and bad! (Especially would appreciate feedback on how well the grouping and scoring performs for you - any quirks?)<p>Thanks!
Nick