Language Prompt for Autonomous Driving

TL;DR

This paper introduces NuPrompt, a large dataset of language prompts for driving scenes, and a Transformer-based model PromptTrack for predicting object trajectories from natural language descriptions, advancing language-guided autonomous driving research.

Contribution

It creates the first object-centric language prompt dataset for driving scenes and formulates a new prompt-based trajectory prediction task with a baseline model.

Findings

PromptTrack achieves impressive performance on NuPrompt.

NuPrompt contains 40,147 language descriptions for driving scenes.

The dataset expands nuScenes with object-centric language annotations.

Abstract

A new trend in the computer vision community is to capture objects of interest following flexible human command represented by a natural language prompt. However, the progress of using language prompts in driving scenarios is stuck in a bottleneck due to the scarcity of paired prompt-instance data. To address this challenge, we propose the first object-centric language prompt set for driving scenes within 3D, multi-view, and multi-frame space, named NuPrompt. It expands nuScenes dataset by constructing a total of 40,147 language descriptions, each referring to an average of 7.4 object tracklets. Based on the object-text pairs from the new benchmark, we formulate a novel prompt-based driving task, \ie, employing a language prompt to predict the described object trajectory across views and frames. Furthermore, we provide a simple end-to-end baseline model based on Transformer, named PromptTrack. Experiments show that our PromptTrack achieves impressive performance on NuPrompt. We hope this work can provide some new insights for the self-driving community. The data and code have been released at https://github.com/wudongming97/Prompt4Driving.