Interpretable Deep Tracking

TL;DR

This paper introduces an interpretable deep learning architecture for multi-object tracking that provides explanations for its decisions by integrating structural causal models and interchange intervention training, enhancing transparency in autonomous vehicle systems.

Contribution

It presents a novel end-to-end trainable multi-object tracking model that incorporates structural causal reasoning for interpretability, addressing the lack of explanations in current deep tracking methods.

Findings

The model can explain its tracking decisions using causal reasoning.

It achieves end-to-end training while maintaining interpretability.

The approach improves understanding of autonomous vehicle decision processes.

Abstract

Imagine experiencing a crash as the passenger of an autonomous vehicle. Wouldn't you want to know why it happened? Current end-to-end optimizable deep neural networks (DNNs) in 3D detection, multi-object tracking, and motion forecasting provide little to no explanations about how they make their decisions. To help bridge this gap, we design an end-to-end optimizable multi-object tracking architecture and training protocol inspired by the recently proposed method of interchange intervention training (IIT). By enumerating different tracking decisions and associated reasoning procedures, we can train individual networks to reason about the possible decisions via IIT. Each network's decisions can be explained by the high-level structural causal model (SCM) it is trained in alignment with. Moreover, our proposed model learns to rank these outcomes, leveraging the promise of deep learning in end-to-end training, while being inherently interpretable.