Mitigating Prior Shape Bias in Point Clouds via Differentiable Center Learning
This addresses a specific bottleneck in point cloud representation learning for computer vision applications, but it is incremental as it builds on existing masked autoencoding methods.
The paper tackles the problem of information leakage in point cloud models due to pre-sampling of center points, which limits learning of global patterns, and introduces DCS-Net to incorporate global and local feature reconstruction, enhancing expressive capacity.
Masked autoencoding and generative pretraining have achieved remarkable success in computer vision and natural language processing, and more recently, they have been extended to the point cloud domain. Nevertheless, existing point cloud models suffer from the issue of information leakage due to the pre-sampling of center points, which leads to trivial proxy tasks for the models. These approaches primarily focus on local feature reconstruction, limiting their ability to capture global patterns within point clouds. In this paper, we argue that the reduced difficulty of pretext tasks hampers the model's capacity to learn expressive representations. To address these limitations, we introduce a novel solution called the Differentiable Center Sampling Network (DCS-Net). It tackles the information leakage problem by incorporating both global feature reconstruction and local feature reconstruction as non-trivial proxy tasks, enabling simultaneous learning of both the global and local patterns within point cloud. Experimental results demonstrate that our method enhances the expressive capacity of existing point cloud models and effectively addresses the issue of information leakage.