KTNet: Knowledge Transfer for Unpaired 3D Shape Completion

Abstract

Unpaired 3D object completion aims to predict a complete 3D shape from an incomplete input without knowing the correspondence between the complete and incomplete shapes. In this paper, we propose the novel KTNet to solve this task from the new perspective of knowledge transfer. KTNet elaborates a teacher-assistant-student network to establish multiple knowledge transfer processes. Specifically, the teacher network takes complete shape as input and learns the knowledge of complete shape. The student network takes the incomplete one as input and restores the corresponding complete shape. And the assistant modules not only help to transfer the knowledge of complete shape from the teacher to the student, but also judge the learning effect of the student network. As a result, KTNet makes use of a more comprehensive understanding to establish the geometric correspondence between complete and incomplete shapes in a perspective of knowledge transfer, which enables more detailed geometric inference for generating high-quality complete shapes. We conduct comprehensive experiments on several datasets, and the results show that our method outperforms previous methods of unpaired point cloud completion by a large margin.

References in corpus (22)

• Distilling the Knowledge in a Neural Network
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
• PF-Net: Point Fractal Network for 3D Point Cloud Completion
• GRNet: Gridding Residual Network for Dense Point Cloud Completion
• ME-PCN: Point Completion Conditioned on Mask Emptiness
• Morphing and Sampling Network for Dense Point Cloud Completion
• 3D Object Reconstruction from a Single Depth View with Adversarial Learning
• PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers
• Shape Inpainting using 3D Generative Adversarial Network and Recurrent Convolutional Networks
• Cascaded Refinement Network for Point Cloud Completion
• You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors
• PMP-Net: Point Cloud Completion by Learning Multi-step Point Moving Paths
• Point Cloud Completion by Skip-attention Network with Hierarchical Folding
• Cycle4Completion: Unpaired Point Cloud Completion using Cycle Transformation with Missing Region Coding
• Unsupervised 3D Shape Completion through GAN Inversion
• Detail Preserved Point Cloud Completion via Separated Feature Aggregation
• ShapeFormer: Transformer-based Shape Completion via Sparse Representation
• LAKe-Net: Topology-Aware Point Cloud Completion by Localizing Aligned Keypoints
• PC${}^2$-PU: Patch Correlation and Point Correlation for Effective Point Cloud Upsampling
• Style-based Point Generator with Adversarial Rendering for Point Cloud Completion
• Learning Local Displacements for Point Cloud Completion
• Learning a Structured Latent Space for Unsupervised Point Cloud Completion

Cited by in corpus (1)

• Weakly Supervised Label Learning Flows