Push-the-Boundary: Boundary-aware Feature Propagation for Semantic Segmentation of 3D Point Clouds
This addresses the challenge of precise object boundary delineation in 3D scene segmentation, which is incremental as it builds on existing feedforward networks with a novel multi-task approach.
The paper tackles the problem of inaccurate semantic segmentation near object boundaries in 3D point clouds by proposing a boundary-aware feature propagation mechanism, resulting in consistent improvements and reduced boundary errors on datasets like S3DIS and SensatUrban.
Feedforward fully convolutional neural networks currently dominate in semantic segmentation of 3D point clouds. Despite their great success, they suffer from the loss of local information at low-level layers, posing significant challenges to accurate scene segmentation and precise object boundary delineation. Prior works either address this issue by post-processing or jointly learn object boundaries to implicitly improve feature encoding of the networks. These approaches often require additional modules which are difficult to integrate into the original architecture. To improve the segmentation near object boundaries, we propose a boundary-aware feature propagation mechanism. This mechanism is achieved by exploiting a multi-task learning framework that aims to explicitly guide the boundaries to their original locations. With one shared encoder, our network outputs (i) boundary localization, (ii) prediction of directions pointing to the object's interior, and (iii) semantic segmentation, in three parallel streams. The predicted boundaries and directions are fused to propagate the learned features to refine the segmentation. We conduct extensive experiments on the S3DIS and SensatUrban datasets against various baseline methods, demonstrating that our proposed approach yields consistent improvements by reducing boundary errors. Our code is available at https://github.com/shenglandu/PushBoundary.