Viewer-Centred Surface Completion for Unsupervised Domain Adaptation in 3D Object Detection

TL;DR

This paper introduces SEE-VCN, a novel viewer-centred surface completion network that enhances 3D object detection across different datasets by focusing on geometry rather than scan pattern overfitting, enabling unsupervised domain adaptation.

Contribution

The paper presents a new viewer-centred surface completion network (VCN) within an unsupervised domain adaptation framework for 3D detection, reducing overfitting to dataset-specific scan patterns.

Findings

Outperforms previous domain adaptation methods in multiple settings

Enables cross-dataset 3D detection without re-training or annotations

Provides a unified, geometry-focused object representation

Abstract

Every autonomous driving dataset has a different configuration of sensors, originating from distinct geographic regions and covering various scenarios. As a result, 3D detectors tend to overfit the datasets they are trained on. This causes a drastic decrease in accuracy when the detectors are trained on one dataset and tested on another. We observe that lidar scan pattern differences form a large component of this reduction in performance. We address this in our approach, SEE-VCN, by designing a novel viewer-centred surface completion network (VCN) to complete the surfaces of objects of interest within an unsupervised domain adaptation framework, SEE. With SEE-VCN, we obtain a unified representation of objects across datasets, allowing the network to focus on learning geometry, rather than overfitting on scan patterns. By adopting a domain-invariant representation, SEE-VCN can be classed as a multi-target domain adaptation approach where no annotations or re-training is required to obtain 3D detections for new scan patterns. Through extensive experiments, we show that our approach outperforms previous domain adaptation methods in multiple domain adaptation settings. Our code and data are available at https://github.com/darrenjkt/SEE-VCN.