|
(0002721)
|
|
Amy Squillacote
|
|
2005-07-25 14:33
|
|
Brad suggested taking the intersection of the bounding box and the view frustum, and use the center of this new volume in calculating the depth of the x-y screen coordinate. There would still be a problem if the current data set was not visible on the screen, but in this case you could use the camera's position as the depth, which would keep you from actually moving.
Another possible solution is to have the user specify/pick a depth value for translation (in perspective projection) similar to setting the center of rotation.
I do not currently have time to implement either solution. |
|