Explainable Action Prediction through Self-Supervision on Scene Graphs
This work addresses action prediction for autonomous driving systems, but it appears incremental as it builds on existing scene graph and self-supervision methods.
The paper tackles the problem of predicting future driver actions in autonomous driving by using scene graphs as a high-level representation, proposing a self-supervision pipeline to handle data scarcity and imbalance, and showing superiority over a fully-supervised approach on the ROAD dataset.
This work explores scene graphs as a distilled representation of high-level information for autonomous driving, applied to future driver-action prediction. Given the scarcity and strong imbalance of data samples, we propose a self-supervision pipeline to infer representative and well-separated embeddings. Key aspects are interpretability and explainability; as such, we embed in our architecture attention mechanisms that can create spatial and temporal heatmaps on the scene graphs. We evaluate our system on the ROAD dataset against a fully-supervised approach, showing the superiority of our training regime.