Evaluation the result & GT pose

NVIDIA-ISAAC-ROS / isaac_ros_visual_slam

Visual SLAM/odometry package based on NVIDIA-accelerated cuVSLAM

https://developer.nvidia.com/isaac-ros-gems

Apache License 2.0

819 stars 127 forks source link

Evaluation the result & GT pose #50

Closed Harin99 closed 1 year ago

Harin99 commented 1 year ago

Thank you for your great work!

And then how did you evaluate your result? Did you get the gt pose in the isaac sim?

I want to use yours using our own dataset generated in isaac sim, we couldn't get the gt pose in isaac sim now. because /tf & /odom is not gt (based on AMCL) and we couldn't find any alternatives.

If you did the evaluation your algorithm, how did you get the GT pose in isaac sim?

Thank you!

swapnesh-wani-nvidia commented 1 year ago

You can just record Isaac Sim data without running VSLAM and collect the ground truth (/tf and /odom) in a ROS bag and use the same bag to evaluate VSLAM. Of course, you have to rename the frame_names either in Isaac Sim or VLSLAM for comparison.