Box2Mask
Box2Mask copied to clipboard
jchibane
Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes
Hi, thanks for sharing this work, seems the training can't stand for several epoches, it can cause oom on a 48GB RAM machine even, seems the code would result in...
Hi, thanks for sharing this great work, I can't run evaluation during training because I don;t have the `gt_instance_data_txt` dir in scannet and seems the script `utils/gt2eval.py` is not working...
First of all, congratulations for your great work. I currently use softgroup to perform instance detection and want to replace it with your work. I only need to make the...
Hi, thanks for this new approach in 3D instance segmentation. I'm interested in replicating your results for the ARKitScene dataset. Would you be able to release a pretrained model on...
Hi, Thanks a lot for your great work. I am really interested in your predictions for the ARKitScenes dataset. Is it possible to release these annotations?
Could you please share how you calculate the superpoints for S3DIS data at link (**oversegmentation** zip file) https://github.com/jchibane/Box2Mask/blob/master/docs/s3dis.md Thank you very much.
I would like to express my appreciation for your notable work in the field of weakly supervised 3D object instance segmentation (3DIS). Your contributions have been significant. I am interested...
Where is the definition of "get_sparse2dense()" called in method Evaluater.s3dis_eval() in file "models/evaluation.py"?
Hello, may I ask how to generate normals files and segment_1abels.tar.gz files if I want to run your model with my own dataset?
