Variational Relational Point Completion Network

2 Knots Observed (3) 1 Knot Observed (2) (4) 2 Knots Observed 0 Knot Observed Complete Shape Our Results Partial Observation Complete Shape Our Results Partial Observation (c) (a) Partial Observation Coarse Completion Fine Completion PCN GRNet NSFA Ours Ground Truth (b) Figure 1: (a) VRCNet performs shape completion with two consecutive stages: probabilistic modeling and relational enhancement. (b) Qualitative Results show that VRCNet generates better shape details than the other works [29, 27, 30]. (c) Our completion results conditioned on partial observations. The arrows indicate the viewing angles. In (1) and (2), 2 knots are partially observed for the pole of the lamp, and hence we generate 2 complete knots. In (3), only 1 knot is observed, and then we reconstruct 1 complete knot. If no knots are observed (see (4)), VRCNet generates a smooth pole without knots.