Robust point-cloud registration based on dense point matching and probabilistic modeling

We present 3D point-cloud registration techniques suited for scenarios where robustness to outliers and missing regions is necessary, besides being applicable both rigid non-rigid configurations. Our exploit advantages from deep learning models dense point matching recent advances in probabilistic modeling of registration. Such a combination produces context awareness resilience information. demonstrate their effectiveness by comparing them state-of-the-art methods showing that ours achieve superior results on existing proposed datasets.