安卓视频识别
我尝试在谷歌上搜索，希望找到任何类似 Google Goggles 的工具/库，在某种程度上允许开发人员识别视频中的对象。
我知道这是一个复杂的话题。
有没有可以识别视频中形状的库？例如检测给定帧/视频中是否有任何方块？我很确定 Goggles 可以做到这一点，但根据我的研究，它没有可供使用的 SDK。
更新：Recognizr 似乎是一部分的一个很好的例子我正在努力实现的目标。但这是一个尚未发布的应用程序。
I tried to google in hope to find any Google Goggles alike tools/libraries that would somewhat allow the developer to recognize objects in a video.
I understand this is a complex topic.
Are there any libraries that would recognize shapes in a video? For instance detect if there are any squares in a given frame/video? I'm pretty sure Goggles can do that, but from what I researched it doesn't have an SDK to use.
UPDATE: Recognizr seems be a good example of a part of what I am trying to accomplish. But it's an application that was not released yet.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
OpenCV（开源计算机视觉）是一个库，很多人都用它来构建这些风格的项目。目前有一个测试质量的项目，可以在 Android 上使用 NDK 运行 OpenCV 代码：
http://opencv.willowgarage .com/wiki/Android
wiki 和项目页面链接了很多 OpenCV 资源，并且很容易找到 OpenCV 的示例。但正如您所说，总体而言，这是一个相当复杂的主题，虽然该框架为您的开发提供了帮助，但在您拥有工作原型之前，仍然需要进行大量分层。
OpenCV (Open Source Computer Vision) is a library that lots of folks use to base these styles of project off of. There's currently a beta-quality project for running the OpenCV code using the NDK on Android:
http://opencv.willowgarage.com/wiki/Android
There are plenty of resources for OpenCV linked to from the wiki and the project page, and it's pretty easy to find samples for OpenCV floating around. But like you said, it's a pretty complex topic overall, and while the framework gives you a leg up on development there will still be plenty to layer in before you even have a working prototype.