aldinorizaldy

The default argument for save_video is False. You should give True to save the video result

Well, I guess in the case of airborne lidar you need to have a large block size. Otherwise misclassification (building roof labeled as ground) will occur on large building as...

Yes, it does make sense. Actually, the default block size in semantic3d code is 5 m. I think you may start with semantic3d code but use your own data, as...

> Thank you for making this code available under MIT license. > > I tried running as-is with latest version of tensor-flow (2.0.0). However, there are major breaking changes between...

@jinghere11 What I did for semantic segmentation is take the partseg model code (Hengshuang) and apply it here https://github.com/yanx27/Pointnet_Pointnet2_pytorch/blob/master/train_semseg.py I only need to change few lines of code

I'm training Hengshuang's model now, but only with ModelNet40 and S3DIS data. I'll let you know when it's done.

> > Thanks for the repo, > > Are there pre-trained model checkpoints available? > > Thanks > > I am looking for the same thing, I even wrote to...

Please download here: https://drive.google.com/file/d/1WxzsRYmhXQDCc4ctLGeVUM0ReBOi9bIC/view?usp=sharing I trained the model for 180 epochs (at that time I miscalculated the training time, so I only got about 180 epochs before my machine shutting...

Hi @kg571852741, I see your dataset is some sort of outdoor scene. Can you elaborate on how you prepare your custom dataset? Thanks in advance.

Hi @Vovinsa, are you able to produce the instance labels for the inference?