Semi-supervised Learning: Fusion of Self-supervised, Supervised Learning, and Multimodal Cues for Tactical Driver Behavior Detection
This work addresses the problem of detecting driver behaviors for automotive safety applications, but it is incremental as it builds on existing methods without claiming major breakthroughs.
The paper tackles tactical driver behavior detection from untrimmed naturalistic driving recordings by addressing challenges like sparse behaviors, long-tail distribution, and intra-class variation, using a fusion of self-supervised and supervised learning with multimodal cues, with preliminary experiments conducted on a 104-hour real-world dataset.
In this paper, we presented a preliminary study for tactical driver behavior detection from untrimmed naturalistic driving recordings. While supervised learning based detection is a common approach, it suffers when labeled data is scarce. Manual annotation is both time-consuming and expensive. To emphasize this problem, we experimented on a 104-hour real-world naturalistic driving dataset with a set of predefined driving behaviors annotated. There are three challenges in the dataset. First, predefined driving behaviors are sparse in a naturalistic driving setting. Second, the distribution of driving behaviors is long-tail. Third, a huge intra-class variation is observed. To address these issues, recent self-supervised and supervised learning and fusion of multimodal cues are leveraged into our architecture design. Preliminary experiments and discussions are reported.