ActiveAnno3D - An Active Learning Framework for Multi-Modal 3D Object Detection

Ahmed Ghita, Bjørk Antoniussen, Walter Zimmer, Ross Greer, Christian Creß, Andreas Møgelmose, Mohan Trivedi, Alois C. Knoll

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

5 Citationer (Scopus)

Abstract

The curation of large-scale datasets is still costly and requires much time and resources. Data is often manually labeled, and the challenge of creating high-quality datasets remains. In this work, we fill the research gap using active learning for multi-modal 3D object detection. We propose ActiveAnno3D, an active learning framework to select data samples for labeling that are of maximum informativeness for training. We explore various continuous training methods and integrate the most efficient method regarding computational demand and detection performance. Furthermore, we perform extensive experiments and ablation studies with BEVFusion and PV-RCNN on the nuScenes and TUM Traffic Intersection (TUMTraf-I) dataset. We show that we can achieve almost the same performance with PV-RCNN and the entropy-based query strategy when using only half of the training data (77.25 mAP compared to 83.50 mAP) of the TUMTraf-I dataset. BEVFusion achieved an mAP of 64.31 when using half of the training data and 52.88 mAP when using the complete nuScenes dataset. We integrate our active learning framework into the proAnno labeling tool to enable AI-assisted data selection and labeling and minimize the labeling costs. Finally, we provide code, weights, and visualization results on our website.
OriginalsprogEngelsk
Titel2024 IEEE Intelligent Vehicles Symposium (IV)
ForlagIEEE (Institute of Electrical and Electronics Engineers)
Publikationsdato2024
ISBN (Trykt)979-8-3503-4882-8
ISBN (Elektronisk)979-8-3503-4881-1
DOI
StatusUdgivet - 2024
BegivenhedIntelligent Vehicles Symposium - Seoul, Sydkorea
Varighed: 2 jun. 20245 jun. 2024
Konferencens nummer: 35

Konference

KonferenceIntelligent Vehicles Symposium
Nummer35
Land/OmrådeSydkorea
BySeoul
Periode02/06/202405/06/2024
NavnI E E E Intelligent Vehicles Symposium
ISSN1931-0587

Fingeraftryk

Dyk ned i forskningsemnerne om 'ActiveAnno3D - An Active Learning Framework for Multi-Modal 3D Object Detection'. Sammen danner de et unikt fingeraftryk.

Citationsformater