A Closer Look at Local Aggregation Operators in Point Cloud Analysis

TL;DR

This paper investigates the impact of local aggregation operators in point cloud networks, finding that many contribute similarly to performance, and introduces a simple, effective operator called PosPool that achieves state-of-the-art results.

Contribution

The paper reveals that various local aggregation operators have similar effects and proposes PosPool, a simple operator that outperforms complex methods in point cloud analysis.

Findings

Different local operators yield similar performance improvements.

PosPool achieves comparable or better results than sophisticated operators.

The proposed method significantly outperforms previous state-of-the-art on PartNet.

Abstract

Recent advances of network architecture for point cloud processing are mainly driven by new designs of local aggregation operators. However, the impact of these operators to network performance is not carefully investigated due to different overall network architecture and implementation details in each solution. Meanwhile, most of operators are only applied in shallow architectures. In this paper, we revisit the representative local aggregation operators and study their performance using the same deep residual architecture. Our investigation reveals that despite the different designs of these operators, all of these operators make surprisingly similar contributions to the network performance under the same network input and feature numbers and result in the state-of-the-art accuracy on standard benchmarks. This finding stimulate us to rethink the necessity of sophisticated design of local aggregation operator for point cloud processing. To this end, we propose a simple local aggregation operator without learnable weights, named Position Pooling (PosPool), which performs similarly or slightly better than existing sophisticated operators. In particular, a simple deep residual network with PosPool layers achieves outstanding performance on all benchmarks, which outperforms the previous state-of-the methods on the challenging PartNet datasets by a large margin (7.4 mIoU). The code is publicly available at https://github.com/zeliu98/CloserLook3D