search

yzqin / s4g-release

S4G: Amodal Single-view Single-Shot SE(3) Grasp Detection in Cluttered Scene
35 stars 5 forks source link

dataset directory #12

Closed xins981 closed 10 months ago

xins981 commented 10 months ago

@yzqin, hi yuzhe. can you provide the full dataset directory structure?

similar below: image

Although I clear up the data_gen code, it's not easy gen training data, I‘sure I'm not alone. this is barrier for promoting this outstanding work. if possible, hope you providing more training details. thank you and your good job!

yzqin commented 10 months ago

hi @xins981

Thank you for your interest in our project. I apologize for any inconvenience, but this work was completed in 2019, which is now 4 years ago. Due to the large size of the dataset file, I had to remove the entire dataset directory, retaining only the code that generates them. The code available here is identical to the one used to generate the training data for the pre-trained model in CoRL 2019.

I regret any trouble this might have caused. After this project, I shifted my research focus and have not worked on grasping detection since, so I do not have any subsequent work related to this paper.

If you have specific technical questions about the code, I'll be more than happy to provide information, to the best of my recollection.

xins981 commented 10 months ago

You're welcome, and thank you for your response. I understand that the work was completed in 2019, and I appreciate your willingness to provide information about the code!