BridgeShape: Latent Diffusion Schr\"odinger Bridge for 3D Shape Completion

ArXi:2506.23205v2 Announce Type: replace Existing diffusion-based 3D shape completion methods typically use a conditional paradigm, injecting incomplete shape information into the denoising network via deep feature interactions (e.g., concatenation, cross-attention) to guide sampling toward complete shapes, often represented by voxel-based distance functions. However, these approaches fail to explicitly model the optimal global transport path, leading to suboptimal completions.