SSeg-LSTM: Semantic Scene Segmentation for Trajectory Prediction

Document Type

Conference Proceeding

Publication Date

8-29-2019

Publication Title

2019 IEEE Intelligent Vehicles Symposium (IV)

First page number:

2504

Last page number:

2509

Abstract

In this paper, we propose the use of semantic segmentation to incorporate scene information for better understanding of human motion in crowded environments. Our proposed SSeg-LSTM method leverages SegNet, which is a semantic segmentation encoder-decoder architecture, to extract semantically meaningful scene features. We then train the Social Scene LSTM (SS-LSTM) model with the contextual information regarding dynamics, social neighborhood, and scene semantics to predict future trajectory points of pedestrians. Experimental evaluation on public datasets show better performance for SSeg-LSTM than SS-LSTM which highlights the utility of semantic encoding for trajectory prediction.

Keywords

Semantic scene segmentation; Trajectory prediction; Scen information; Human motion; Crowded environment; SS-LSTM; LSTM model

Disciplines

Electrical and Computer Engineering | Engineering

Language

English

UNLV article access

Search your library

Share

COinS