Gating Syn-to-Real Knowledge for Pedestrian Crossing Prediction in Safe Driving

TL;DR

This paper introduces a gated knowledge transfer framework for pedestrian crossing prediction that effectively leverages synthetic data and multiple domain adaptation techniques to improve real-world prediction accuracy.

Contribution

It proposes a novel gated knowledge fusion method combining style transfer, distribution approximation, and knowledge distillation for diverse domain knowledge adaptation in PCP.

Findings

Achieved superior performance on real datasets PIE and JAAD.

Constructed a large synthetic benchmark S2R-PCP-3181 with diverse data.

Demonstrated effective knowledge transfer from synthetic to real-world scenarios.

Abstract

Pedestrian Crossing Prediction (PCP) in driving scenes plays a critical role in ensuring the safe operation of intelligent vehicles. Due to the limited observations of pedestrian crossing behaviors in typical situations, recent studies have begun to leverage synthetic data with flexible variation to boost prediction performance, employing domain adaptation frameworks. However, different domain knowledge has distinct cross-domain distribution gaps, which necessitates suitable domain knowledge adaption ways for PCP tasks. In this work, we propose a Gated Syn-to-Real Knowledge transfer approach for PCP (Gated-S2R-PCP), which has two aims: 1) designing the suitable domain adaptation ways for different kinds of crossing-domain knowledge, and 2) transferring suitable knowledge for specific situations with gated knowledge fusion. Specifically, we design a framework that contains three domain adaption methods including style transfer, distribution approximation, and knowledge distillation for various information, such as visual, semantic, depth, location, etc. A Learnable Gated Unit (LGU) is employed to fuse suitable cross-domain knowledge to boost pedestrian crossing prediction. We construct a new synthetic benchmark S2R-PCP-3181 with 3181 sequences (489,740 frames) which contains the pedestrian locations, RGB frames, semantic images, and depth images. With the synthetic S2R-PCP-3181, we transfer the knowledge to two real challenging datasets of PIE and JAAD, and superior PCP performance is obtained to the state-of-the-art methods.