The Mixed Reality Forums here are no longer being used or maintained.
There are a few other places we would like to direct you to for support, both from Microsoft and from the community.
The first way we want to connect with you is our mixed reality developer program, which you can sign up for at https://aka.ms/IWantMR.
For technical questions, please use Stack Overflow, and tag your questions using either hololens or windows-mixed-reality.
If you want to join in discussions, please do so in the HoloDevelopers Slack, which you can join by going to https://aka.ms/holodevelopers, or in our Microsoft Tech Communities forums at https://techcommunity.microsoft.com/t5/mixed-reality/ct-p/MicrosoftMixedReality.
And always feel free to hit us up on Twitter @MxdRealityDev.
Can I create a building of real size when using Hololens out door?
For example, when I use HoloLens on the street or square, can I create an Eiffel Tower model of real size(300 meters high)? Assuming that model is simplified, no performance issue, and also ignore the fov of HoloLens.
If I can, how far I can walk around to see the Eiffel Tower in the right position? If I walk 100 meters away, can I see it?
Best Answers
-
Options
mark_grossnickle
✭✭✭
I believe you can create the Eiffel Tower given the constraints you just listed.
As for how far away... I am not sure. I know you can edit the camera's rendering range by modifying the clipping planes on the camera. You can set it to 100 meters but I am not sure about the spatial mapping at that distance or whether or not the spatial mapping could track everything in the area between you and the tower (if that is important to your use case).
Taqtile
5 -
Optionsmark_grossnickle
✭✭✭
I believe the hololens knows the spatial mapping of the room the are in, where the hololens is located within that spatial map, and which direction the hololens is pointed at.
From there it does a raycast and attempts to hit a collider. That collider could be something created in Unity or it could be the spatial mapping if it is enabled.
So I think the big question in your case is can you place the building and then walk 100 meters away and stay in the same room/spatial mapping. I'd venture to guess there would be issues there.
Taqtile
5 -
Options
trzy
✭✭✭
The HoloLens does not use the spatial mesh for real-time position tracking. I'm not sure exactly how the algorithm works but my understanding is that it has multiple (4?) optical image sensors specifically for this purpose. Presumably, it uses some fancy computer vision algorithms for identifying features and tracking them. This is why it is said to track better in a "cluttered" room than a featureless one.
To answer Steven's question about how hot spots are created in, say, a museum application, I think this is usually done with the assistance of WorldAnchors. My understanding is that these also are not pinned to spatial mesh data and presumably contain identifying information from the optical sensors (perhaps in combination with depth data) that allow the HoloLens to correlate the anchor with features in the room.
Bart
http://trzy.org2 -
Options
thebanjomatic
✭✭✭
The one thing you will have to be careful with when using World Anchors is rotational errors due to the lever-arm effect. If you are placing the large building using a world anchor in the center, then tiny little position updates in that coordinate system can result in large sweeping movement far away from the anchor.
Basically if the anchor rotates by only half a degree to correct for error, there isn't that much of a change close to the anchor itself. On the other hand, a point 100m away will rotate along an larger arc. In terms of straight-line distances, that point will be 87cm away from its starting location.
As a result, world anchors probably aren't the best solution for placing a large building (or at least not in terms of placing them with a single anchor).
https://developer.microsoft.com/en-us/windows/mixed-reality/coordinate_systems
6
Answers
I believe you can create the Eiffel Tower given the constraints you just listed.
As for how far away... I am not sure. I know you can edit the camera's rendering range by modifying the clipping planes on the camera. You can set it to 100 meters but I am not sure about the spatial mapping at that distance or whether or not the spatial mapping could track everything in the area between you and the tower (if that is important to your use case).
Taqtile
Thank you:)
I am wandering how HoloLens locate the position of player/user? Without laser station like HTC-Vive, or any GPS device, how HoloLens know the user just moved certain range and direction?
In some museum projects, when user gaze at a picture or a dinosaur skeleton, according actions are triggered, but I want to know how HoloLens knows the user gaze the right target? Is it implemented by image recognition? or HoloLens just knows the position of user and triggered the according logic?
I believe the hololens knows the spatial mapping of the room the are in, where the hololens is located within that spatial map, and which direction the hololens is pointed at.
From there it does a raycast and attempts to hit a collider. That collider could be something created in Unity or it could be the spatial mapping if it is enabled.
So I think the big question in your case is can you place the building and then walk 100 meters away and stay in the same room/spatial mapping. I'd venture to guess there would be issues there.
Taqtile
Thank you very much.
When I get the HoloLens, I will make a test on it and reply on the thread:)
The HoloLens does not use the spatial mesh for real-time position tracking. I'm not sure exactly how the algorithm works but my understanding is that it has multiple (4?) optical image sensors specifically for this purpose. Presumably, it uses some fancy computer vision algorithms for identifying features and tracking them. This is why it is said to track better in a "cluttered" room than a featureless one.
To answer Steven's question about how hot spots are created in, say, a museum application, I think this is usually done with the assistance of WorldAnchors. My understanding is that these also are not pinned to spatial mesh data and presumably contain identifying information from the optical sensors (perhaps in combination with depth data) that allow the HoloLens to correlate the anchor with features in the room.
Bart
http://trzy.org
WorldAnchors? Thank you for giving me such a keyword, is it a C# function/class in Unity coding? Although I do not get HoloLens yet, from the function name I know positions of users and objects can be stored in a world coordinate, even it is a black box for me by now.
In a video, the man created a screen in the house, then he went out of the house and walked in the garden, through the window we can even see the screen in the room at the right position, that is very cool. In the end he return his room and the screen is still there. I think that is what I need.
World Anchor? Thank you for giving me the keyword. Is it an function or class that can be referenced in Unity coding? From the name I know positions of user and objects can be tracked in a world coordinate, even it is just a black box for me by now.
In this video, the man created a screen in the room, then walked out. Through the window we even can see the screen. He walked in the garden for a while and back to the room, the screen was still there in the same position, I think that is what I need, but I do not know if he restart the HoloLens, the screen can be in the same position? If it can, that will be very cool.
Thank you for giving me the key word "World Anchor".
In this video a man created a screen and walked out, we can see the screen is still in the right position, I think that is what I want.
The one thing you will have to be careful with when using World Anchors is rotational errors due to the lever-arm effect. If you are placing the large building using a world anchor in the center, then tiny little position updates in that coordinate system can result in large sweeping movement far away from the anchor.
Basically if the anchor rotates by only half a degree to correct for error, there isn't that much of a change close to the anchor itself. On the other hand, a point 100m away will rotate along an larger arc. In terms of straight-line distances, that point will be 87cm away from its starting location.
As a result, world anchors probably aren't the best solution for placing a large building (or at least not in terms of placing them with a single anchor).
https://developer.microsoft.com/en-us/windows/mixed-reality/coordinate_systems
Thank you for your answer and document link.
You give me an important key point, the rotation. In my draft, take Louvre Museum for example, user only to adjust my model Venus to overlap the real one for once, the anchor was made, so all other positions like Mona Lisa was coordinated because I measured the vector between Venus and Mona Lisa and stored in the program, but I forget the important thing, the rotation deviation.
For museum projects, it is import to know "Where am I" and "What I am looking at". The HoloLens can scan A sculpture in front of me for every detail, but can not know Which sculpture is in front of me, especially after restart the program, because it seems do not have an absolute coordinate system like GPS, and 2D image recognition is not so powerful by now,
When the user wearing HoloLens, walking in front the sculpture of Venus, how does HoloLens know that is Venus and trigger according logic?
Thank you for your answer and document link, I will read through it.
For museum projects, it is important to know "Where am I?" and "What I am looking at?". For example, when user wearing HoloLens and staying in front of the sculpture of Venus, the HoloLens can scan that there is a sculpture there, but how does it know that is Venus? especially after restart the program.
In my previous draft, I will let user to adjust my model Venus to overlap the real one for one time, I thought after that all other arts are located because I measured vectors of all artworks in the museum from itself to Venus, but I forget the import thing, the rotation deviation.
WorldAnchors? Thank you for giving me such a keyword, is it a C# function/class in Unity coding? Although I do not get HoloLens yet, from the function name I know positions of users and objects can be stored in a world coordinate, even it is a black box for me by now.
In a video, the man created a screen in the house, then he went out of the house and walked in the garden, though the window we can even see the screen in the room at the right position, that is very cool. In the end he return his room and the screen is still there. I think that is what I need.
WorldAnchors? Thank you for giving me such a keyword, is it a C# function/class in Unity coding? Although I do not get HoloLens yet, from the function name I know positions of users and objects can be stored in a world coordinate, even it is a black box for me by now.
In a video, the man created a screen in the house, then he went out of the house and walked in the garden, though the window we can even see the screen in the room at the right position, that is very cool. In the end he return his room and the screen is still there. I think that is what I need.