Language-Driven Active Learning for Diverse Open-Set 3D Object Detection

Ross Greer, Bjørk Antoniussen, Andreas Møgelmose, Mohan Trivedi

Keywords: vision-language models, autonomous driving, active learning, 3D object detection

TL;DR: Diversity-based active learning techniques rely on a measurement of what makes data novel relative to points which exist in the training set; we show that language-based image encodings offer a measure of diversity which benefits active learning.

Abstract:

Object detection is crucial for ensuring safe autonomous driving. However, data-driven approaches face challenges when encountering minority or novel objects in the 3D driving scene. In this paper, we propose VisLED, a language-driven active learning framework for diverse open-set 3D Object Detection. Our method leverages active learning techniques to query diverse and informative data samples from an unlabeled pool, enhancing the model’s ability to detect underrepresented or novel objects. Specifically, we introduce the Vision-Language Embedding Diversity Querying (VisLED-Querying) algorithm, which operates in both open-world exploring and closed-world mining
settings. In open-world exploring, VisLED-Querying selects data points most novel relative to existing data, while in closed-world mining, it mines new instances of known classes. We evaluate our approach on the nuScenes dataset and demonstrate its effectiveness compared to random sampling and entropy-querying methods. Our results show that VisLED-Querying consistently outperforms random sampling and offers competitive performance compared to entropy-querying despite the latter’s model-optimality, highlighting the potential of VisLED for improving object detection in autonomous driving scenarios. We make our code publicly available at [anonymized].

Submission Number: 20