Desktop uploader

The Mapillary Desktop Uploader is a desktop application that enables uploading Mapillary imagery in bulk. You can upload to individual as well as organization accounts (including both public and private organization imagery).

The Desktop Uploader is available for macOS and Windows and currently supports two types of file formats: geotagged image files and BlackVue dashcam recordings. We're hoping to support more imagery and video types in the future (including 360° imagery that needs stitching).

Download the Desktop Uploader →

In this article, you can read about:

1. Getting started
2. Uploading images
3. Uploading BlackVue dashcam recordings
4. Upload history and retrying uploads
5. Aborting the upload and reprocessing images
6. Reporting issues (including where to find error logs)

Note. To ensure our service is fast and high-performing, we've set an upper limit of 100 MP to image resolution. If your imagery exceeds that, please reduce the resolution before uploading, otherwise it won't get processed.

 

Getting started

Download the installer file from the download page and run it on your computer like when installing any other application.

Open it and sign in with your Mapillary login credentials. You can’t use the application without signing in, so if you don’t have an account yet, create one on the Mapillary web.

Once you’ve signed in, you’re ready to start uploading.

1. Choose your upload target
2. Add files (they will be processed for uploading)
3. Upload

 

Uploading images

After you've signed in to the uploader, the first step is to make sure you've chosen the correct upload target.

If you are a member of an organization on Mapillary, you will see the option to change your upload target (individual or organization, public or private imagery). Do that before you add imagery (if you forget, you need to reprocess the images before uploading—read below how to do that).

Drag and drop or browse for a folder that contains your images (to avoid errors, please make sure the folder name doesn't contain special characters). Note that you can’t add individual images, just folders. You can only add one folder at a time. However, the folders can contain subfolders—the uploader is capable of handling that.

Your imagery needs to be already geotagged (i.e. have GPS information in the EXIF data); you can't add that to images in the Uploader. In the future, we hope to add options for processing and editing imagery pre-upload.

Note. If you want to upload images captured with the Mapillary mobile apps, you need to take some extra steps to make sure the images copied to your computer are geotagged. For iOS, see the instructions here. For Android, you need to use designated software to add location data to images from the separately saved GPX file.

 

When you’ve added a folder, the images it contains will be pre-processed and you will see their locations displayed on the map. Depending on the volume, this may take some time. At this stage, the images are not yet uploaded. Note that if you’re trying to add a folder that contains images which have already been processed previously, the uploader will remember that and prompt you to resume the upload (even if you've added new images to the folder that are not processed).

When the imagery has been processed, you can preview it by clicking on an image marker on the map and then the image icon on the right-hand side. You can also see image details if you click on the button in the viewer that says EXIF.

If you want, you can edit the compass angle of the whole batch before you upload. Click on “Advanced options” and enter the offset angle in degrees (0 for north, 90 for east, 180 for south, 270 for west).

Finally, click “Upload” to start uploading the prepared images to Mapillary. You can leave your computer; the application will prevent it from going to sleep before it has completed the upload. This way, you can easily opt to upload images at the time you’re not using your bandwidth for other things (e.g. overnight).

 

Uploading BlackVue dashcam recordings

The Desktop Uploader supports the front camera video from the Blackvue DR900s (1-channel and 2-channel) models.

To upload videos, first make sure you've chosen the correct upload target in the drop-down menu of the file selection screen.

Next, add the folder(s) containing the videos. The uploader will prepare them for uploading (this may take a while, depending on the file volume). You'll see the GPS tracks of the processed videos on the map.

Hit the Upload button at the bottom left. You can leave the upload unattended; the application will prevent your computer from going to sleep until all files have been uploaded. If your connection drops, the Uploader will retry upload a number of times automatically. If repeatedly unsuccessful, the upload will be listed in your Upload history so you can retry manually later (see the next section).

After uploading, Mapillary will sample your videos for image frames. You should see your processed uploads on the platform within some hours normally (note that processing videos takes a bit longer compared to if you upload images).

 

Upload history and retrying uploads

At the top right of the Uploader window, you can see a button for accessing your upload history. There you’ll see all your uploads—completed as well as failed ones. This will help you keep track of what you’ve managed to upload. You can see which account you uploaded to, how many images or videos, when, and whether the upload was successful. For failed or canceled uploads, you can click the “Retry” button right there to resume the upload.

From the three dots at the end of the row, you can access some additional options such as opening the folder on your computer that contains the files, deleting the files from your computer, or removing this upload from the uploads list.

 

Aborting the upload and reprocessing images

The current version of the Desktop Uploader doesn't include editing options and expects that you have prepared your imagery before adding the folder to the uploader. If you discover in the processing screen that you still need to change something (e.g. remove certain images) or forgot to choose the correct upload target, you need to reprocess the imagery.

As mentioned earlier, the uploader will remember if you’ve already processed a folder and prompt you to resume the upload. To prevent that and be able to reprocess the imagery, you need to delete the log files that were created during the previous processing.

Go to the folder (or subfolders) in your Finder (macOS) or Explorer (Windows) that contain the images and delete the hidden folder called .mapillary. To see it, you need to toggle viewing hidden files:

• Hold down `Cmd` + `Shift + `.` (period) in the Finder on macOS
• Change the folder settings on Windows—see here for instructions for different OS versions

After deleting the .mapillary folder, you can add the image folder to the uploader again to reprocess the contents (instead of resuming the upload).

Note. If you've processed the imagery previously with our command line tools, you may not be able to upload it successfully with the Desktop Uploader. We recommend that you complete the upload with command line tools as well in that case. This is a known issue that we'll try to address in the future.

 

Error and bug reporting

If you run into any errors with the Uploader, please email us at support@mapillary.com and provide the following information:

• Description of the problem (what happened vs what did you expect)
• Platform (macOS or Windows)
• App version (from the "About" menu item)
• A detailed description of the steps taken
• Screenshots or videos illustrating the problem
• Logfiles

To find the logfiles (you may need to allow viewing hidden files, see the previous section):

• On macOS: ~/Library/Logs/<app name>/log.log
  For example: ~/Library/Logs/Mapillary\ Uploader/log.log (expanded to /Users/username/Library/Logs/Mapillary\ Uploader/log.log)
• On Windows: %USERPROFILE%\AppData\Roaming\<app name>\log.log
  For example: C:\Users\IEUser\AppData\Roaming\Mapillary Uploader\log.log