The Best of Both Worlds: Combining CNNs and
Geometric Constraints for Hierarchical Motion Segmentation
Abstract
Traditional methods of motion segmentation use powerful geometric constraints to understand motion, but fail
to leverage the semantics of high-level image understanding. Modern CNN methods of motion analysis, on the
other hand, excel at identifying well-known structures, but
may not precisely characterize well-known geometric constraints. In this work, we build a new statistical model
of rigid motion flow based on classical perspective projection constraints. We then combine piecewise rigid motions
into complex deformable and articulated objects, guided by
semantic segmentation from CNNs and a second “objectlevel” statistical model. This combination of classical geometric knowledge combined with the pattern recognition
abilities of CNNs yields excellent performance on a wide
range of motion segmentation benchmarks, from complex
geometric scenes to camouflaged animals