Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

T265 Known Issues and Robustness #7258

Open
cimbar opened this issue Sep 3, 2020 · 8 comments
Open

T265 Known Issues and Robustness #7258

cimbar opened this issue Sep 3, 2020 · 8 comments

Comments

@cimbar
Copy link

cimbar commented Sep 3, 2020

Hi,
There are 3 main known issues about T265 and it stays there at each new release.

Known Issues
#4261 - [T265] Add ability to open multiple devices from different processes.
#4518 – [T265] Pose data produces NaNs. Can still occur in some cases. If detected, please attempt to make a raw data (images + IMU) recording using the recorder tool, and attach a link to it in the github issue, to assist our resolution.
[T265][Mac] - Start after stop is not working on Mac with the T265 camera

Is realsense developer team working on these known issues? It looks that there is not much interest on fixing T265 issues. This camera developed for robotics application but it is still not possible to develop robust system with this camera since it was released. Users always face an issue about connection, stability or accuracy. I wonder that is there a roadmap about these known issues and will there be major update in soon?

@mpottinger
Copy link

For me it is accuracy and relocalization that is the problem. In a handheld application it has nowhere near as good tracking stability and recovery as ARCore and ARKit despite having two fisheye cameras dedicated to tracking.

@stevemartinov
Copy link

I gave up on this camera while ago, this is an absolute piece of crap. It constantly gave me wrong readings (usually -0.5m offset within 10 meter range) and relocalization/pose_jumping simply do not work (even though it can jump to a correct position but all previous nodes are not optimized and therefore right after the jump, if you move backwards to n-1 node, it will jump back again! What a crap!)

@mpottinger
Copy link

Iand relocalization/pose_jumping simply do not work

Yes this is a problem for me. It does not work like on mobile phone AR SDK's, where if it loses tracking it jumps back to the correct pose relatively quickly. If the T265 goes off, it doesn't recover very easily/quickly at all, sometimes not at all. Meanwhile I could wave my ARCore phone around like crazy and pose will recover extremely quickly and reliably.

@msadowski
Copy link

According to the reply I've received here it doesn't seem like there will be any work done on T265, which is a pity as the idea behind this sensor was so interesting!

@RealSenseSupport
Copy link
Collaborator

T265 is not discontinued or retired, and is available. We have moved our focus to our next generation of products and consequently, we will not be addressing issues/new features in the T265.

@arjunsengupta98
Copy link

According to the reply I've received here it doesn't seem like there will be any work done on T265, which is a pity as the idea behind this sensor was so interesting!

Agreed :(

@lishanggui
Copy link

lishanggui commented Sep 23, 2020

Hi @RealSenseSupport my t265 issue at #7199. can any people help me? thanks!

The sensor of T265 (including IMU) is calibrated on the production line, so no further calibration process is required (unlike the IMU on D435i). However, realsense-viewer shows that the acceleration will vary from 8 to 10 when placed at rest in different directions. Does this have an impact on vio and how to solve it?

@zhengzh
Copy link

zhengzh commented Oct 14, 2021

@RealSenseSupport we just bought three T265 camera

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

8 participants