3D pose estimation is within the scope of WikiProject Robotics, which aims to build a comprehensive and detailed guide to Robotics on Wikipedia. If you would like to participate, you can choose to edit this article, or visit the project page (Talk), where you can join the project and see a list of open tasks.
Are there any cases in which the term "pose estimation" could refer to "2D pose estimation"? I can't think of any. kostmo (talk) 03:16, 3 March 2010 (UTC)
2D pose can normally be thought of as simply aligning two corresponding data sets through an homography -- while this is strictly true for 3D pose, the calculation (well estimation) of the homographies in each case - 2D & 3D - are significantly different. Much work in the realm of 2D has been completed, while that of 3D currently remains sparse with much of the head way being purely at University publication level; having not yet achieved mainstream text-book publication.
Currently the article only mentions 3D from 2D extraction, there are one or two methods fro 3D-to-3D pose transformations that should also be added into this article, to allow a 3D to 3D homography.
I would suggest labelling the article a stub and requesting expansion.
One thing that needs to be taken into account if a merging is done is that THIS article appears to be about a specific pose estimation method whereas THAT article tries to make a more general overview of the area. Also, I believe that pose, in terms of position and orientation of an object, is relevant also for the 2D case, where then normally only the position and orientation relative to the image frame can be determined. Seach "2D pose estimation" on Google and you'll find plenty of publications on 2D pose and its estimation. Personally, I'd rather see a separation between, on the one hand, a general article on pose as a concept used in computer vision, how and why it used, different strategies for pose estimation, and, on the other hand, specific methods/algorithms for estimating pose. I guess that as long as there aren't sufficiently much stuff in either of the two types of articles (like it is now) a merging makes sense, but as soon as the general part and/or the specific methods expand too much there will have to be a split again. --KYN (talk) 09:16, 3 March 2010 (UTC)
I would like to remove the whole "3d pose estimation" article. It summarizes in a bad and missleading way some of my former research.
"Pose estimation" has different definitions. I prefer a definition given by Grimson (1981) ( ... the transformation to map an object model from its inherent coordinate system into agreement with sensor data). This is independent from the dimension and can be applied in 1D,2D, 2D-3D, 3D ...
oppose merge There are actually two separate topics, one in robotics, another in vision, and neither topic has been treated well. Both topics need to be fixed, not merged while half-done. History2007 (talk) 10:57, 6 February 2011 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.