Compass direction or heading on the Ricoh Theta V where is it?


#1

Today I received my Ricoh Theta V and I produced some of my first shots. Once I uploaded two files from my Android device via the Google StreetView app I noticed that the initial orientation that is shown in the app does not match the direction that is visually presented to the user as rotated image. I was under the impression that the Ricoh Theta V had a compass build in. Using exiv2 I am enable to recover the following information:

exiv2 -p a R0010002_20181008184158.JPG  | grep -i Pose
Xmp.GPano.PosePitchDegrees                   XmpText     4  -0.9
Xmp.GPano.PoseRollDegrees                    XmpText     4  -0.6

Exiftool shows something like below. It is pretty obvious that the -0.6 is actually the 359.43 - 360.

xiftool -Accelerometer R0010002_20181008184158.JPG 
Accelerometer                   : 359.43 -0.9

On other places on this forum I find that people talk about an InitialViewHeadingDegrees / PoseHeadingDegrees. What can I expect from this device? Does this device know the correct orientation or is only a gyroscope available and not a magnetic compass? If a compass is not available what is a best practise to add orientation to a single exposure, can this be done via the API?


#2

I installed a compass app from here

It seems to work running directly on the THETA V. I compared the orientation to another compass and it seems accurate. The screenshot below is from Vysor displaying the THETA V.

I’m looking at the Exif info using ExifToolGUI and I think the compass information is stored into the image at GPSImgDirection. I’m guessing, but that’s what it looks like to me.

image


#3

Hi @codetricity thanks for your reply. I am going to try to get my camera in developer mode (and see how people do these Android tricks). I am very surprised your camera actually reports a compass direction. I do hope that the animation is not caused by the gyroscope and/or the cellphone input. Using exiv2 -p a on my image I am not receiving GPSImgDirection in my output.

Update: I have installed AndroSensors on the device via adb install. I can confirm I indeed see a magnetic field and orientation output which does something when I move the device. But the values also jump over 10 degrees while the device is fully static. I am going to make another shot, and see if using ExifToolGUI I am able to retreiceve GPSImgDirectionRef.

Update 2: Magic, after enabling developer mode and using AndroSensors I am receiving:

Exif.GPSInfo.GPSImgDirectionRef Ascii 2 True direction
Exif.GPSInfo.GPSImgDirection Rational 1 13570/100

I have checked the photo’s I made in the morning, they do not contain direction. Maybe it would be a good idea to inform Ricoh about this issue. Today I hand positioned all my Streetview contributions. If only I knew.

Update 3: Reply from Ricoh:

Thank you for using RICOH THETA Partner Program.
Regarding questions from you, we will answer as follows.

The electronic compass dynamically changes whether direction information cannot be created in the image depending on whether calibration was possible or not.

When you move the camera closer to the magnet, direction information will not be attached to the image.
After that, turn the camera around and after a while, the calibration will be performed and direction information will be attached to the image.


#4

Wow, this is great information. thank you for posting it. This is very helpful for other people.

the vague part is “after a while, the calibration will be performed”. We can probably do more tests to clarify what that line means.

Does the default mobile app attach the compass heading to GPSImgDirection?

Thanks.


#5

Does the default mobile app attach the compass heading to GPSImgDirection?

I wrote a very elaborate e-mail to them with virtually all options tested. So “the button” press adds direction (if available), the app adds GPS (if available from the mobile phone) and compass from the Theta (if available). Now the Theta has been switched on and off multiple times, while no additional calibration was performed, still something we must figure out when it looses this calibration.

For a first time user I think we should add a FAQ section that the user should attach a wrist wrap and rotate their camera over every axis while it is on.


#6

NOTE: Inaccurate information I found in a wrong manual deleted.

I think someone else on this forum had a similar problem with compass calibration recently.


#7

I think someone else on this forum had a similar problem with compass calibration recently.

You should see the number of posts on the localguides fora regarding compass offsets. I wonder if it is possible to get my device in a state of ‘uncalibrated’ I have enough magnets around me :wink: Then it would be interesting to make some tutorial.


#8

I have recently bought a complete new camera. This entirely new camera has the same "issue" as mentioned before.

In the documentation something is written as:
"If the lamp is lit in red, the electromagnetic compass needs to be calibrated. Move te camera in a figure eight until the lamp turns blue."

I would like to emphase that this description does not happen. There is no red light, and the photos created with the camera have no orientation. After the camera is rotated over all axis when it is turned on it indeed works as expected, with orientation in Exif.


#9

I think you mentioned the manual for the 1st generation “RICOH THETA” on 2013, not for “RICOH THETA V”.


#10

Nice catch.

This is the link for the THETA V.

https://support.theta360.com/en/manual/v/content/shooting-photo/shooting_photo_01.html