search

dougsm / ggcnn

Generative Grasping CNN from "Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach" (RSS 2018)
BSD 3-Clause "New" or "Revised" License
499 stars 140 forks source link

How to visualise with 3D graspping pose? #54

Open yananliusdu opened 1 year ago

yananliusdu commented 1 year ago

Hi Doug,

thanks for your amazing work, I managed to run your project and get some results (Figure1) from cornell dataset and Jacquard dataset.

My question is how to get the visualisation as shown in your paper (Figure2) ?

thanks

Figure 1 Figure_1

Figure 2 Screenshot from 2023-03-09 11-10-08

dougsm commented 1 year ago

Hey there,

The second visualisation is from RViz (the visualisation tool from ROS). You can see an example of how it's published here. But converting the GG-CNN output to a 6D Pose requires the camera intrinsics (see here), so it may not be possible just from the Cornell or Jacquard datasets. Hope that helps.