This paper was about using a touch screen to move objects in a three dimensional space using multiple contact points in direct manipulation. It discussed several aspects of this including world space transformations, minimization methods, three-finger rotations, ambiguous rotations, rotational exhaustion, and more. The main content was focused on how to move the object with particular finger movements such as using two contact points to signify an axis, and using another finger to rotate about that axis. Another example would be using one finger to select a point on the object, and moving that finger to directly translate the object. Two fingers could be used to do things such as scaling the object by moving on finger closer/further away from the other, translating similar to the one finger movement, and so on. The direct manipulation aspect meant that the contact points were directly mapped to the object, so when the fingers moved those points on the object moved in that fashion. To compensate for the various movements a user could make the object would either translate, rotate, scale, or a combination of the three in order to achieve the desired result. The paper discussed the authors experience with the system and problems they encountered such as ambiguous transformations and rotational exhaustion, as well as methods to minimize the affects of these like using pressure, biasing, and many others.
Here is an image of some three finger movements:
Discussion:
This paper turned out to be pretty interesting because I have some basic experience in graphics so I understood most of what was being discussed such as the projection and orthogonal views, the use of quaternions, and the transformations such as scaling, translating, rotating, shear, etc. The direct manipulation of the finger contact points to the object was pretty interesting because I think that when dealing with this situation that is the most intuitive way users would want to move the objects because as it said in the paper it lets them feel as if they are 'gripping' the object. I really liked how the author discussed his first experience with the system and the problems he encountered. The ambiguous rotations was pretty interesting because it would be something I wouldn't have thought of. I also liked that some of the ways they treated these problems were more like patches since technically the system behaved correctly just not as the user would have expected. I'm not really sure how this could be advanced for further use, but I liked the idea of using it for moving around a landscape as some of the images showed. I think using pressure add another dimension to the system, allowing the user and alternative movement to make the object move.
No comments:
Post a Comment