将人类转化为 3D
这可能与 stackoverflow 不完全相关，但我没有更好的网站来发布此内容。
我们收到了开发某种能够让真人进入包括纹理的 3D 应用程序的请求。
现在我不知道如何处理这个问题，因为我们以前没有经验。那么这可能吗？如果可能的话我们需要什么样的东西？
编辑：只是一些附加信息，这是我学校的一位老师提出的一个想法，他希望它能够在几分钟内扫描人员。因此，最可行的方法似乎是使用多个摄像头可能与绿/蓝屏幕结合使用。
This might not be completely relevant to stackoverflow but I have no better site where to post this.
We have gotten a request to develop something that can get a real human into a 3d application including textures.
Now I have no idea how to handle this as we have no previous experience. So is this possible and if it's possible what kind of things would we need ?
Edit: Just some additional info, this is an idea a teacher at my school came up with and he wants it to be able to scan people in, in a few minutes. So the most viable thing seems to be the use of multiple camera's possibly in combination with a green/blue screen.
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（3）
如果您需要纹理，则至少需要一台相机。您可能需要安装多个摄像头并使用立体视觉技术来获取 3D 信息。您还可以将一台相机与一台或多台激光扫描仪或飞行时间相机的测量融合在一起（产生距离图像）。
如果您还需要运动模型，您应该看看体育游戏的功能（EA Sports、Pro Evolution Soccer...）。据我所知，他们根据人们身上的标记来记录动作。
也许这有助于作为第一个开始。
If you need textures, you'll need at least one camera. You might want to install several cameras and use stereo vision techniques to get the 3d information. You could also fuse one camera with measurements of one or more laser scanners or Time of Flight Cameras (yielding distance images).
If you also need motion models you should take a look at what sports games do (EA Sports, Pro Evolution Soccer...). As far as I know they record motion based on marks placed on people.
Maybe this helps as a first start.
苏格兰格拉斯哥有一家公司在这方面拥有丰富的经验：
Dimensional Imaging Ltd.
他们的演示看起来令人印象深刻，我知道格拉斯哥已经对此进行了很多原创研究。
There is a company in Glasgow, Scotland that have a lot of experience with this:
Dimensional Imaging Ltd.
Their demos look impressive and I know a lot of original research on this has been done in Glasgow.
原则上，您需要使用 3D 扫描仪来捕获人的 3D 结构。我对这些没有经验，但它们是商业上可用的。
对于表面，您可以只拍摄人物，然后使用软件创建纹理（您需要消除透视畸变等）。然后在 3D 软件中组装所有内容。
这只是一个非常的粗略轮廓；这个过程相当复杂。
如果我问：你是如何在“没有经验”的情况下完成如此复杂的项目的？看起来很不寻常...
In principle you need to use a 3D scanner to capture the 3D structure of the person. I have no experience with these, but they are commercially available.
For the surface, you could just photograph the person, then use software to create textures (you'd need to remove perspective distortion etc.). Then assemble everything in a 3D software.
That's just a very rough outline; the process is rather complex.
If I may ask: How did you get such a complex project with "no previous experience"? Seems rather unusual...