Enhancing Performance of Point Cloud Completion Networks with Consistency Loss

2024-10-09

arXiv.org

AI Summary

Point cloud completion suffers from conflicting supervision signals and training instability due to the inherent one-to-many mapping between a single incomplete input and multiple plausible completions. To address this, we propose the first Completion Consistency Loss for point cloud completion, which enforces semantic consistency among completions generated by the network from different masked versions of the same source point cloud—thereby mitigating the one-to-many ambiguity. This loss is architecture-agnostic: it requires no modification to the backbone network, integrates seamlessly with standard reconstruction losses (e.g., Chamfer Distance), and leverages data augmentation to construct co-source masked sample pairs. Evaluated on benchmarks including MVP, our method consistently improves multiple state-of-the-art (SOTA) models, achieving new SOTA performance without incurring any additional inference overhead.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

Point cloud completion networks are conventionally trained to minimize the disparities between the completed point cloud and the ground-truth counterpart. However, an incomplete object-level point cloud can have multiple valid completion solutions when it is examined in isolation. This one-to-many mapping issue can cause contradictory supervision signals to the network because the loss function may produce different values for identical input-output pairs of the network. In many cases, this issue could adversely affect the network optimization process. In this work, we propose to enhance the conventional learning objective using a novel completion consistency loss to mitigate the one-to-many mapping problem. Specifically, the proposed consistency loss ensure that a point cloud completion network generates a coherent completion solution for incomplete objects originating from the same source point cloud. Experimental results across multiple well-established datasets and benchmarks demonstrated the proposed completion consistency loss have excellent capability to enhance the completion performance of various existing networks without any modification to the design of the networks. The proposed consistency loss enhances the performance of the point completion network without affecting the inference speed, thereby increasing the accuracy of point cloud completion. Notably, a state-of-the-art point completion network trained with the proposed consistency loss can achieve state-of-the-art accuracy on the challenging new MVP dataset. The code and result of experiment various point completion models using proposed consistency loss will be available at: https://github.com/kaist-avelab/ConsistencyLoss .

Problem

• Point Cloud Completion
• Learning Instability
• Consistency Issues

Innovation

• Completion Consistency Loss
• Point Cloud Completion
• Accuracy Enhancement