Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

NYC-Event-VPR: A Large-Scale High-Resolution Event-Based Visual Place Recognition Dataset in Dense Urban Environments

Taiyi Pan, Junyang He, Chao Chen, Yiming Li, Chen Feng

TL;DR

The NYC-Event-VPR dataset is introduced to the robotics and computer vision communities, featuring the Prophesee IMX636 HD event sensor (1280x720 resolution), combined with RGB camera and GPS module, and employs three frameworks to conduct generalization performance assessments, promoting innovation in event-based VPR and its integration into robotics applications.

Abstract

Visual place recognition (VPR) enables autonomous robots to identify previously visited locations, which contributes to tasks like simultaneous localization and mapping (SLAM). VPR faces challenges such as accurate image neighbor retrieval and appearance change in scenery. Event cameras, also known as dynamic vision sensors, are a new sensor modality for VPR and offer a promising solution to the challenges with their unique attributes: high temporal resolution (1MHz clock), ultra-low latency (in μs), and high dynamic range (>120dB). These attributes make event cameras less susceptible to motion blur and more robust in variable lighting conditions, making them suitable for addressing VPR challenges. However, the scarcity of event-based VPR datasets, partly due to the novelty and cost of event cameras, hampers their adoption. To fill this data gap, our paper introduces the NYC-Event-VPR dataset to the robotics and computer vision communities, featuring the Prophesee IMX636 HD event sensor (1280x720 resolution), combined with RGB camera and GPS module. It encompasses over 13 hours of geotagged event data, spanning 260 kilometers across New York City, covering diverse lighting and weather conditions, day/night scenarios, and multiple visits to various locations. Furthermore, our paper employs three frameworks to conduct generalization performance assessments, promoting innovation in event-based VPR and its integration into robotics applications.

NYC-Event-VPR: A Large-Scale High-Resolution Event-Based Visual Place Recognition Dataset in Dense Urban Environments

TL;DR

The NYC-Event-VPR dataset is introduced to the robotics and computer vision communities, featuring the Prophesee IMX636 HD event sensor (1280x720 resolution), combined with RGB camera and GPS module, and employs three frameworks to conduct generalization performance assessments, promoting innovation in event-based VPR and its integration into robotics applications.

Abstract

Visual place recognition (VPR) enables autonomous robots to identify previously visited locations, which contributes to tasks like simultaneous localization and mapping (SLAM). VPR faces challenges such as accurate image neighbor retrieval and appearance change in scenery. Event cameras, also known as dynamic vision sensors, are a new sensor modality for VPR and offer a promising solution to the challenges with their unique attributes: high temporal resolution (1MHz clock), ultra-low latency (in μs), and high dynamic range (>120dB). These attributes make event cameras less susceptible to motion blur and more robust in variable lighting conditions, making them suitable for addressing VPR challenges. However, the scarcity of event-based VPR datasets, partly due to the novelty and cost of event cameras, hampers their adoption. To fill this data gap, our paper introduces the NYC-Event-VPR dataset to the robotics and computer vision communities, featuring the Prophesee IMX636 HD event sensor (1280x720 resolution), combined with RGB camera and GPS module. It encompasses over 13 hours of geotagged event data, spanning 260 kilometers across New York City, covering diverse lighting and weather conditions, day/night scenarios, and multiple visits to various locations. Furthermore, our paper employs three frameworks to conduct generalization performance assessments, promoting innovation in event-based VPR and its integration into robotics applications.

Paper Structure

This paper contains 10 sections, 1 equation, 7 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. NYC-Event-VPR
  4. Sensors
  5. Data Collection
  6. Benchmark Experiments
  7. Experiment setup
  8. Benchmark findings
  9. Conclusion and Impact
  10. Future Directions

Figures (7)

  • Figure 1: Downtown Manhattan in the event frame, reconstructed frame, and RGB frame.
  • Figure 2: NYC-Event-VPR covers NYC, focusing on Chinatown area in Manhattan with overlapping traversal. Each color represents a distinct traversal.
  • Figure 3: Our sensor suite: (a) Prophesee EVK4 HD Event Camera, (b) ELP USB Camera, (c) Insta360 vibration damper, (d) Sparkfun GPS-RTK-SMA GPS module
  • Figure 4: Sensor setup and mounting design: RGB camera is mounted on top of event camera, and the sensor suite is positioned facing forward behind vehicle's front windshield
  • Figure 5: Example images in processed dataset (from left to right columns): naive conversion, E2VID reconstruction, RGB reference. Each row is the same visual scene. Each column is the same dataset.
  • ...and 2 more figures