Class-aware tiny object recognition over large-scale 3D point clouds

Jialin Li

doi:10.1016/j.neucom.2023.01.094

Return

Class-aware tiny object recognition over large-scale 3D point clouds

2023-04-01

PRE

DOI：10.1016/j.neucom.2023.01.094

Original

Original request for help

Save

Abstract

En 中文

Although tremendous strides have been made in object recognition over large-scale 3D point clouds, one of the remaining open challenges is detecting tiny objects. The signal and appearance information of tiny objects are generally sparse and insufficient due to their tiny size, which makes the tiny object difficult to be detected with existing studies both in tiny 2D object recognition and general 3D object recognition since 3D topology relations are lost due to feature projection in tiny 2D object recognition and massive background may dominate the feature learner in general 3D object recognition. To the best of our knowl-edge, we explore tiny object recognition for the first time over large-scale 3D point clouds by designing deep neural network. To cope with this problem, a novel two-stage approach is proposed in this paper, which can effectively learn an informative and discriminative feature representation and efficiently pro-cess large-scale point clouds. Specifically, a class-aware filtering strategy is designed to dispense with the redundant background information. After that, we introduce a novel aggregating and sampling scheme in a supervised manner to progressively increase the receptive field for each 3D point, which can adaptively capture useful signals from tiny objects and learn complex geometric structures. Last, we propose a biased loss function that is inversely correlated to the number of tiny object points, thereby effectively handling the sparsity of tiny objects. Extensive experimental results over two real-world data sets vali-date the effectiveness and efficiency of the proposed approach.(c) 2023 Elsevier B.V. All rights reserved.

Keywords:

Tiny object recognition

Point cloud analysis

Signal sparsity

Deep neural network