This paper presents an optimization framework for estimating the motion and underlying physical parameters of a rigid body in free flight from video. The algorithm takes a video clip of a tumbling rigid body of known shape and generates a physical simulation of the object observed in the video clip. This solution is found by optimizing the simulation parameters to best match the motion observed...

The estimation of structure and motion from image sequences is one of the most studied problems within computer vision. However, almost all the efforts in this area have dealt with rigid objects. Our surrounding environment is generally not a rigid place, with swaying trees, moving people and rolling waves. Hence, to have structure and motion systems work effectively in our world, in general, t...

We introduce a novel multiframe scene flow approach that jointly optimizes the consistency of the patch appearances and their local rigid motions from RGB-D image sequences. In contrast to the competing methods, we take advantage of an oversegmentation of the reference frame and robust optimization techniques. We formulate scene flow recovery as a global non-linear least squares problem which i...

A direction-length framework is a pair (G, p), where G = (V ;D,L) is a ‘mixed’ graph whose edges are labeled as ‘direction’ or ‘length’ edges, and p is a map from V to R for some d. The label of an edge uv represents a direction or length constraint between p(u) and p(v). Let G be obtained from G by adding, for each length edge e of G, a direction edge with the same end vertices as e. We show t...

This paper presents the experimental validation of an approach based on a coordinate-free representation to recognize six DOF rigid body motion trajectories. In this approach, the three dimensional measured position trajectories of arbitrary and uncalibrated points attached to the rigid body are transformed to an invariant, coordinatefree representation of the rigid body motion trajectory. This...