DIY Streetview for Krpano Panoramic Viewer

This repository contains Python tools for linking panoramic images into a self-hosted virtual tour, a-la Google Streetview and generating XML configuration files for the Krpano panoramic viewer. These tools allow you to create a custom Streetview-like experience by connecting and displaying your panoramic images. Currently, the only tested camera model is Ricoh Theta S. If you want to use it with the images, captured by any other camera, you will need to edit Python scripts to change image resolution and names of the XMP image tags in the convert_panoramas.py. You're welcome to open an issue on GitHub if you need help with that.

Scripts

convert_panoramas.py: This script generates a point shapefile from a set of panoramic images using their geolocation data and performs geometric correction of the images using Nona from Hugin tools.

panoramic_calc.py: This script connects the panoramic images based on a specified radius or a supplied line shapefile, connecting the location of the images. It also creates XML configuration files for the Krpano panoramic viewer containing metadata for each panoramic image point, such as the connected neighboring points and their azimuths.

Test shapefiles

The test_shapefiles directory contains two example shape files.

• points.shp - this was file created by convert_panoramas.pyscript from the set of panoramic images.
• lines.shp - this is an example line shape file with the links between the images. It was created by connecting the image locations from points.shp in QGIS. You might want to use these files as an imput to the visualize_connections.py script to learn how it works.

Viewer template

The viewer_template folder contains KRPano viewer configuration files, that you need to publish on your webserver along with the XML files, created by the panoramic_calc.py. KRPano is a commercial software, and you will need to add tour.js file from it to the viewer_template folder to be able to use this solution. You might want to use a KRPano demo version for the initial test. It will display a watermark.

Requirements

To run these scripts, you'll need Python 3.x and the following packages:

geopandas matplotlib pandas PyQt5 shapely You can install them using pip:

pip install geopandas matplotlib pandas PyQt5 shapely

Usage

Prior to using the scripts, you will need a set of images, collected by the Ricoh Theta S panoramic camera. The images should have GPS location stored in their EXIF tag. It means that you have to collect the images when the camera is connected to the smartphone with the Ricoh Theta S application running, or use a Python script running on your smartphone to trigger the camera and update the GPS location.

1. Create Point Shapefile and apply geometric correction to the images

Run create_point_shapefile.py to generate a point shapefile using the geolocation data from your set of panoramic images. You'll need to provide the script with the appropriate input parameters (e.g., input image file folder, output image folder, location to save the output point shapefile).

2. Visualize Connections and Generate Krpano XML Configuration Files

Run the panoramic_calc.py script. A graphical user interface (GUI) will open, allowing you to:

Specify whether to use a search radius or line shapefile for determining connections between panoramic images. Select the point shapefile created using create_point_shapefile.py. If not using a search radius, select the line shapefile, that you prepaired using QGIS or any other GIS software package. Select an output folder where the Krpano XML configuration files will be saved. Click "Run" to visualize the connections and generate the Krpano XML configuration files. The generated Krpano XML configuration files will contain the following information for each point:

• Point name
• Names of connected neighboring points
• Azimuths of the connections
• The connections between the points will be visualized in the GUI.

Integration with Krpano Panoramic Viewer

• Create a virtual tour in the KRPano by dropping your input images to MAKE VTOUR (NORMAL) droplet.bat.
• Make a copy of the viewer_template folder to a convenient location
• Copy the tour.js file from the KRPano output folder to the copy of the viewer_template folder.
• Copy the created image scenes from your tour to the panos folder in the copy of the viewer_template folder.
• Copy the xml files, created by the panoramic_calc.py to the scenes folder in the copy of the viewer_template folder.
• Start your local testing web server or copy the contents to the web server. Navigate to view.html?s=. KRPano will display you a specific panoramic image. You can then navigate between the scenes using hotspots, double click or using W,A,S,D keys on the keyboard.
• You might want to create a web map, allowing the users to open the links to view.html?s= to view different panoramas. See the example on using ArcGIS Online at https://github.com/ngolosov/CollegeCourtStreetview You could implement a web map using any open source web mappling API of your choice, such as Leaflet or OpenLayers.

License

This project is licensed under the MIT License