DROID: Driver-centric Risk Object Identification

TL;DR

This paper introduces DROID, a novel framework for identifying objects that influence driver behavior using egocentric video, aiming to improve subjective risk assessment in driving scenarios.

Contribution

The paper presents a new cause-effect based task and a two-stage framework for driver-centric risk object identification using egocentric video data.

Findings

Achieved state-of-the-art performance on HDD dataset

Validated the effectiveness of the two-stage DROID framework

Demonstrated DROID's applicability for risk assessment

Abstract

Identification of high-risk driving situations is generally approached through collision risk estimation or accident pattern recognition. In this work, we approach the problem from the perspective of subjective risk. We operationalize subjective risk assessment by predicting driver behavior changes and identifying the cause of changes. To this end, we introduce a new task called driver-centric risk object identification (DROID), which uses egocentric video to identify object(s) influencing a driver's behavior, given only the driver's response as the supervision signal. We formulate the task as a cause-effect problem and present a novel two-stage DROID framework, taking inspiration from models of situation awareness and causal inference. A subset of data constructed from the Honda Research Institute Driving Dataset (HDD) is used to evaluate DROID. We demonstrate state-of-the-art DROID performance, even compared with strong baseline models using this dataset. Additionally, we conduct extensive ablative studies to justify our design choices. Moreover, we demonstrate the applicability of DROID for risk assessment.