GoPro Omni Upload to Google Street View

In this walkthrough you will learn how to use the GoPro Omni to capture and stitch panoramic images. As well as publishing those images into the Google Street View app.

 

Workflow for uploading to street view

I have made a YouTube walkthrough of how to do this. You can watch that below and or read the text as well.

 

Photos taken on GoPro Omni

We need to do go and take photos of our subject for the Google Street View upload. In this case I have walked around a local park and taken a lot of images using the GoPro Omni. With all these images, we are going to modify them and stitch them together.  

 

Note : Before I made this tutorial I was using the GoPro Omni Importer v2.5. This version did not include the feature to importing all your photos.  So you couldn’t automatically get each micro SD card into Autopano Pro. But now the process is a lot easier on v2.6. I will go through both ways in case you are using an older firmware and Importer software.

File organization (GoPro Omni Importer 2.5 or Lower)

Image of gopro omni file structure to import into autopano pro

First thing we need to do is copy over all of our files from each GoPro micro SD card and onto our computer hard drive. I have an external hard drive specifically to hold all my files. Next we want to organize them so it is easier to import each photo to stitch together into the final panoramic image.

 

I like to group up each of the 6 GoPro Hero 4 files for each panoramic shot because it is easier to organize and use in future edits.

Image of gopro omni grouping files for easy import

 

Once we are done creating groups for each panoramic image, we can start to import each of these grouped images into Autopano Pro.

Import into Autopano Pro (GoPro Omni Importer v2.5 or lower)

So now that we have all our folders nice and organized, we can easily import all the images into a group on the Autopano interface. Pressing Detect at the top left will pre-process the panoramic in the sample on the right side.

Image of kolor autopano pro software automatically grouping imported images

 

We can now see what it will look like, sometimes the preview doesn’t align the horizon correctly so you might have to click and drag around on the panoramic preview to fix that.

Once you have realigned the horizon, we can make more changes by clicking on the edit button, we can render it or queue it up for batch rendering to be ran in the future.

Image of kolor autopano pro software feature to batch render images at a later time

I am going to go ahead and  batch render a few so we can get some complete panoramic images.

 

Import into Autopano Giga with GoPro Omni Importer v2.6

When we use the Omni Importer v2.6 or greater it gives us the ability to skip the previous steps of organizing the files and manually creating and importing groups from those folders.

Image of gopro omni importer options to process and auto open in kolor autopano pro software

The Omni Importer v2.6 gives us the option to just process the file and automatically open up the processed group into Autopano Pro.  You can continue on through the steps of scheduling rendering each image using the queue functionality.

 

ExifTool to inject metadata

Now that we have successfully rendered and exported our panoramic images, we need to inject metadata into the images to make sure all 360 viewers understand that it is an equirectangular image and not a basic photo from our phone. You can download the correct ExifTool here (http://www.sno.phy.queensu.ca/~phil/exiftool/).

 

Image of a meta data injecting tool for exif data

 

Make sure to follow the instructions for the specific operating system your computer has. There are system requirements per OS if you scroll down on the site from the link provided above.

Go through the directions for using the tool on the website and inject each panoramic image.

 

Image of command line executions for a exif tool

Once you are done, we can now clean up the shadows, tripods and or monopod from the final image.

 

Using Photoshop and Flexify to remove shadows and Tripods

Now that we have all of our images injected with the metadata that we need, we can now update each one and remove any unwanted details for each image.

Download Adobe Photoshop and make sure to instal the Flexify 2 add on library. Both have trial periods that you can use to see how well they work. Find and follow instructions to install Flexify 2 depending on your operating system.

Once we successfully installed both, open up one panoramic image with a tripod, shadow or person you want to remove from the photo.

First thing we want to do is go over to the right and duplicate the current background layer.

 

Image of how to duplicate a layer in adobe photoshop software

 

Next we want to a filter onto one of our layers. Go to the top of the screen in photoshop and click on the filter tab, scroll down to Flaming Pear and click on Flexify 2

 

Image of the location for a filter option called flaming pear flexify two for the adobe photoshop software

 

We are now on the flaming pear add on that enables us to do a lot of fancy filters to our layer. What we want to do is change the current view on the image from equirectangular to the nadir. So go down to where a dropdown says Input and select ‘equirectangular’ and then where a dropdown says output and select ‘zenith & nadir’. It will do some pre-processing and display a preview on the right side. You should now see a very clear view of the tripod, shadow or person holding a monopod.

 

Image of the options to select in the flexify filter

 

After applying the filter to the layer it will take some time to buffer and process the filter. Now we have the image where we want it. Use the stamp and blend tools to slowly remove the object from the final image.

 

Image of the stamp and blend tools in adobe photoshop

 

Once you have finished your stamping and blending on the layer, we can now go back and apply the exact opposite filter that we did the first time. Go back to the filter tool, down to flaming pear and select flexify two. This time, we select input to be zenith & nadir and output to be equirectangular.

 

Image of the flexify plug in options for input and output

Once we have updated our layer with the newest filter, we can combine the layers. But, we need to do it in a specific way. Select both layers, right click and select merge visible.

 

Image of merging two layers using the merge visible option in adobe photoshop

 

We now have a perfect panoramic image without any distractions. Next we can go through this process with every single image that we took so we can create a very high quality Google Street View with actionable hotspots to travel around the park.

 

Upload the images into Google Street View App

Now that we have all of our images for the park, we can upload them to Google Drive. Once we have done that, open up Google Drive on your phone.

 

Image of google drive on a phone

 

Once in the app find your images and select all and share to the Google Street View App. Open up the Google Street View app and select your profile to view the most recently updated or shared images.

 

Switch to the private tab and you will see all the photos that are available to be published. Click on the ‘Add a Maps Listing’ to the group of images. Type in the location of park that we just took pictures of.

 

Image of the google streetview app and assigning a gps location as well as moving and connecting views to complete the google street view process

 

Next we need to connect the images and place hotspots, so click select all and move and connect photos.

Publish and Connect Google Street View Images

Once the loading is complete, you will see a bunch of dots and lines connecting them all. Each one is an image and you need to find the correct location on the map to place them. Once you place them in the correct spots, we can link up their neighbors. Click on each node and connect the lines to their neighbor nodes. Doing this will automatically create hotspots for us to travel from one spot to the other.

 

Image of the completed google streetview connected up and ready to be publish for the public to view

 

After taking some time to connect all of the dots and lines, we have a finished map. Typically you don’t want too many lines connecting to each dot or node. Because, the user might have trouble clicking on the correct hotspot to move to the next view. Finally, click the checkbox and accept the changes, now it will take a while for your street view to be public.