Transferable Class Statistics and Multi-scale Feature Approximation for 3D Object Detection

TL;DR

This paper proposes a lightweight 3D object detection method from point clouds using multi-scale feature approximation via knowledge distillation, transferable class-aware statistics, and a new localization metric, achieving effective results with reduced computational cost.

Contribution

It introduces a novel approach to approximate multi-scale features from a single neighborhood and employs class-aware statistics for transferable features, reducing computational complexity.

Findings

Effective 3D detection with less computation

Improved localization accuracy using central weighted IoU

Validated on public datasets with strong results

Abstract

This paper investigates multi-scale feature approximation and transferable features for object detection from point clouds. Multi-scale features are critical for object detection from point clouds. However, multi-scale feature learning usually involves multiple neighborhood searches and scale-aware layers, which can hinder efforts to achieve lightweight models and may not be conducive to research constrained by limited computational resources. This paper approximates point-based multi-scale features from a single neighborhood based on knowledge distillation. To compensate for the loss of constructive diversity in a single neighborhood, this paper designs a transferable feature embedding mechanism. Specifically, class-aware statistics are employed as transferable features given the small computational cost. In addition, this paper introduces the central weighted intersection over union for localization to alleviate the misalignment brought by the center offset in optimization. Note that the method presented in this paper saves computational costs. Extensive experiments on public datasets demonstrate the effectiveness of the proposed method.