Abstract:

Many non-rigid 3D structures are not modelled well through a low-rank subspace assumption. This is problematic when it comes to their reconstruction through Structure from Motion (SfM). We argue in this paper that a more expressive and general assumption can be made around compressible 3D structures. The vision community, however, has hitherto struggled to formulate effective strategies for recovering such structures after projection without the aid of additional priors (e.g. temporal ordering, rigid substructures, etc.). In this paper we present a "prior-less" approach to solve compressible SfM. Specifically, we demonstrate how the problem of SfM - assuming compressible 3D structures - can be theoretically characterized as a block sparse dictionary learning problem. We validate our approach experimentally by demonstrating reconstructions of 3D structures that are intractable using current state-of-the-art low-rank SfM approaches.

Results:

Red dots: The proposed method. Blue circles: Ground Truth. Green dots: Dai et.al.'s method. The proposed method obtained an impressive performance for compressible structures. However it fails in Shark sequences due to the poor coherence of learned dictionary.

Code

Visit GitHub for MATLAB code.

Publication

  • Prior-less Compressible Structure from Motion[pdf].
    Chen Kong, Simon Lucey.
    In Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, Nevada, USA, June 2016.