search

HITSZ-NRSL / Dynamic-VINS

[RA-L 2022] RGB-D Inertial Odometry for a Resource-restricted Robot in Dynamic Environments
310 stars 40 forks source link

ground truth of HITSZ & THUSZ Datasets #39

Closed Mahmoudadham4040 closed 3 months ago

Mahmoudadham4040 commented 7 months ago

could you please provide us with the ground truth or the RTK-GPS trajectories -if any- for the HITSZ & THUSZ Datasets ?

jianhengLiu commented 7 months ago

Sorry that these datasets do not contain GT or GPS trajectory. It is collected for qualitative experiments, and you could compare results with loop-closure trajectory or satellite map.

Mahmoudadham4040 commented 7 months ago

Thanks for your consideration. I have another point, I don't have depth image messages and want to evaluate Vins with yolo? could you please provide me your suggestions to edit in this code?

jianhengLiu commented 7 months ago

One plausible way is to modify codes to add choice to receive datastream without depth, and mask all features in yolo bbox

---Original--- From: @.> Date: Tue, Nov 14, 2023 19:25 PM To: @.>; Cc: @.**@.>; Subject: Re: [HITSZ-NRSL/Dynamic-VINS] ground truth of HITSZ & THUSZ Datasets(Issue #39)

Thanks for your consideration. I have another point, I don't have depth image messages and want to evaluate Vins with yolo? could you please provide me your suggestions to edit in this code?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>