Chrome Extension
WeChat Mini Program
Use on ChatGLM

Vision-Based Navigation in Indoor Environments Without Using Image Database

PROCEEDINGS OF THE 27TH INTERNATIONAL TECHNICAL MEETING OF THE SATELLITE DIVISION OF THE INSTITUTE OF NAVIGATION (ION GNSS 2014)(2014)

Korea Inst Sci & Technol

Cited 0|Views2
Abstract
Recently, an indoor navigation has attracted great attention in industries and research institutes. Also, a wearable device such as smart glasses has been emerging as the next-generation device to substitute the smartphone. We have a plan to develop an indoor navigation system based on smart glasses as soon as it is released in Republic of Korea. One of the most critical problems in vision-based indoor navigation is difficulty in the database construction when estimating the locations of users.Here, we propose a vision-based navigation system in indoor environments which uses an indoor map and does not require an image database. In each image, features are extracted to recognize the indoor objects without using image database. We define the metric about the four features which are pillar detection metric (PDM), hallway entrance detection metric (HEDM), hallway detection metric (HDM), and absence. Since the four metrics are normalized in the range of 0 or 1, we can classify them using the decision tree. After the classification, we estimate the position of the user through the angle-based matching between the recognized objects and the objects in an indoor map. We already know the angular information of each image using magnetometer and gyroscope built in smartphone. So, we can know the angle between the current position and recognized object. The current position of user is represented as a linear equation and we can solve it using pseudo inverse.To verify the performance of the proposed system, we conducted the positioning experiments in indoor environment. The indoor map is comprised of hall and hallway and we captured the image in six positions which were four halls and two hallways. In each position, we captured 18 images around the user with its angular information. The experimental results show an average recognition rate with 71.3% and the positioning performance with 7 meters.
More
Translated text
求助PDF
上传PDF
Bibtex
AI Read Science
AI Summary
AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.
Example
Background
Key content
Introduction
Methods
Results
Related work
Fund
Key content
  • Pretraining has recently greatly promoted the development of natural language processing (NLP)
  • We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
  • We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
  • The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
  • Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance
Upload PDF to Generate Summary
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Data Disclaimer
The page data are from open Internet sources, cooperative publishers and automatic analysis results through AI technology. We do not make any commitments and guarantees for the validity, accuracy, correctness, reliability, completeness and timeliness of the page data. If you have any questions, please contact us by email: report@aminer.cn
Chat Paper
GPU is busy, summary generation fails
Rerequest