Search papers, labs, and topics across Lattice.
This paper explores the use of rolling-shutter distortions in cameras to estimate the 3D translational and angular velocities of rapidly moving spherical objects, turning a common artifact into a valuable source of information. By designing a detectable spherical pattern and employing a correspondence-free back-projection framework, the authors achieve decoupled motion estimation through a two-stage optimization process. Extensive experiments validate the method's robustness and accuracy in high-speed scenarios, even with textureless objects.
Rolling-shutter distortions can be harnessed to achieve accurate velocity estimates of fast-moving spheres, even in the absence of texture.
Rolling-shutter cameras introduce characteristic distortions when imaging fast moving objects, and these effects are typically treated as artifacts to be corrected. In this work, we instead leverage rolling-shutter distortions as a valuable source of temporal information to estimate the 3D translational and angular velocities of rapidly moving spherical objects from a single rolling-shutter frame. We design a robust and easily detectable spherical pattern and propose a correspondence-free formulation that recovers motion by enforcing geometric consistency in a back-projection framework. By exploiting the geometry of the sphere, translational and rotational motions are decoupled and estimated through a two-stage optimization process, enabling reliable velocity recovery even for textureless objects. Extensive experiments on both synthetic and real datasets demonstrate accurate and robust estimation of motion parameters under challenging high-speed conditions.
