PointNCBW: Towards Dataset Ownership Verification for Point Clouds via Negative Clean-label Backdoor Watermark

TL;DR

This paper introduces PointNCBW, a novel dataset watermarking method for point clouds that enables ownership verification by embedding stealthy backdoor triggers across all classes, resistant to removal, and effective in black-box settings.

Contribution

The paper proposes a scalable clean-label backdoor watermarking scheme for point cloud datasets that works across all classes and enhances dataset ownership verification.

Findings

Effective watermarking across all classes in large-scale datasets

High resistance to removal and detection methods

Successful verification in black-box model settings

Abstract

Recently, point clouds have been widely used in computer vision, whereas their collection is time-consuming and expensive. As such, point cloud datasets are the valuable intellectual property of their owners and deserve protection. To detect and prevent unauthorized use of these datasets, especially for commercial or open-sourced ones that cannot be sold again or used commercially without permission, we intend to identify whether a suspicious third-party model is trained on our protected dataset under the black-box setting. We achieve this goal by designing a scalable clean-label backdoor-based dataset watermark for point clouds that ensures both effectiveness and stealthiness. Unlike existing clean-label watermark schemes, which are susceptible to the number of categories, our method could watermark samples from all classes instead of only from the target one. Accordingly, it can still preserve high effectiveness even on large-scale datasets with many classes. Specifically, we perturb selected point clouds with non-target categories in both shape-wise and point-wise manners before inserting trigger patterns without changing their labels. The features of perturbed samples are similar to those of benign samples from the target class. As such, models trained on the watermarked dataset will have a distinctive yet stealthy backdoor behavior, i.e., misclassifying samples from the target class whenever triggers appear, since the trained DNNs will treat the inserted trigger pattern as a signal to deny predicting the target label. We also design a hypothesis-test-guided dataset ownership verification based on the proposed watermark. Extensive experiments on benchmark datasets are conducted, verifying the effectiveness of our method and its resistance to potential removal methods.