0

I have a very specific application in which I would like to try structure from motion to get a 3D representation. For now, all the software/code samples I have found for structure from motion are like this: "A fixed object that is photographed from all angle to create the 3D". This is not my case.

In my case, the camera is moving in the middle of a corridor and looking forward. Sometimes, the camera can look on other direction (Left, right, top, down). The camera will never go back or look back, it always move forward. Since the corridor is small, almost everything is visible (no hidden spot). The corridor can be very long sometimes.

I have tried this software and it doesn't work in my particular case (but it's fantastic with normal use). Does anybody can suggest me a library/software/tools/paper that could target my specific needs? Or did you ever needed to implement something like that? Any help is welcome!

Thanks!

Jean-François Côté
  • 4,200
  • 11
  • 52
  • 88

2 Answers2

4

What kind of corridors are you talking about and what kind of precision are you aiming for?

A priori, I don't see why your corridor would not be a fixed object photographed from different angles. The quality of your reconstruction might suffer if you only look forward and you can't get many different views of the scene, but standard methods should still work. Are you sure that the programs you used aren't failing because of your picture quality, arrangement or other reasons?

If you have to do the reconstruction yourself, I would start by
1) Calibrating your camera
2) Undistorting your images
3) Matching feature points in subsequent image pairs
4) Extracting a 3D point cloud for each image pair

You can then orient the point clouds with respect to one another, for example via ICP between two subsequent clouds. More sophisticated methods might not yield much difference if you don't have any closed loops in your dataset (as your camera is only moving forward).

OpenCV and the Point Cloud Library should be everything you need for these steps. Visualization might be more of a hassle, but the pretty pictures are what you pay for in commercial software after all.

Edit (2017/8): I haven't worked on this in the meantime, but I feel like this answer is missing some pieces. If I had to answer it today, I would definitely suggest looking into the keyword monocular SLAM, which has recently seen a lot of activity, not least because of drones with cameras. Notably, LSD-SLAM is open source and may not be as vulnerable to feature-deprived views, as it operates directly on the intensity. There even seem to be approaches combining inertial/odometry sensors with the image matching algorithms.

Good luck!

FvD
  • 1,286
  • 13
  • 25
  • I am not using a stereo camera, I am using structure from motion. Stereo camera would be very nice but the video are given by the client from different kind of camera. – Jean-François Côté May 07 '13 at 12:29
  • You don't need a stereo setup to match features in image pairs, and in fact that's what the SFM demo software you linked does. Looking for correspondences in your images is at the base of both structure from motion and stereo vision. What about the other questions? Do you know the reasons for other algorithms failing in your case? – FvD May 07 '13 at 21:37
  • I think it's because of a lack of texture/feature to recognize. It's kind of a linear pattern and I think it fail to match. Also, since everything is view from the same angle, the model created is realy distorted. I will try again with visual SFM tomorrow . – Jean-François Côté May 07 '13 at 23:44
  • No luck again with the Visual SfM. It seems that it doesn't work because of the texture and the fact that it comes from a video. Thanks for you help! – Jean-François Côté May 13 '13 at 11:41
1

FvD is right in the sense that your corridor is a static object. Your scenario is the same and moving around and object and taking images from multiple views. Your views are just not arranged to provide a 360 degree view of the object.

I see you mentioned in your previous comment that the data is coming from a video? In that case, the problem could very well be the camera calibration. A camera calibration tells the SfM algorithm about the internal parameters of the camera (focal length, principal point, lens distortion etc.) In the absence of knowledge about these, the bundler in VSfM uses information from the EXIF data of the image. However, I don't think video stores any EXIF information (not a 100% sure). As a result, I think the entire algorithm is running with bad focal length information and cannot solve for the orientation.

Can you extract a few frames from the video and see if there is any EXIF information?

Mustafa
  • 367
  • 1
  • 4
  • 13
  • there are no EXIF information in these frame because it's extracted from a video like you said (with openCV). – Jean-François Côté May 13 '13 at 11:40
  • If you still have access to the camera that took the video, you can try calibrating the camera yourself by taking a video of a calibration pattern and then manually entering the camera information in a text file for VSfM. – Mustafa May 23 '13 at 23:55
  • unfortunately, the video are given by the client (so we don't know the camera) and made by hundreds of different models... But thanks for the idea! – Jean-François Côté May 27 '13 at 11:35