Charles R. Qi

Hi @themmes thanks for the comments. @keczub you are very welcomed to try the model on outdoor dataset, but the exact performance will depend on many things I'd say like...

The question has been a while.. The motivation to add X'Y'Z' is because for PointNet the model is applied to individual rooms. The normalized coordinates of points in a room...

Hi @luoxiaoliaolan Surely you are working on some important problem. There are some challenges on how to segment large-scale scenes like outdoor ones scanned by LiDAR like how to capture...

You are welcomed to refer here: https://github.com/charlesq34/3dmodel_feature/blob/master/io/write_hdf5.py for how to prepare HDF5 files. Best, Charles

Hi @karenachiketc If it is a classification problem: yes, but you also need to change the model definition file for size of the output layer, and train.py for the `num_classes`...

Hi @karenachiketc The steps sound good to me. Just make sure you visualize the point clouds before you start training, to make sure they are correctly prepared. You can either...

I think one can easily convert those data types to raw point clouds as N by C arrays. You are welcomed to post your converting code (link) here if you'd...

HI @brbrhpt As there are batch norm layers, using batch size = 1 would cause some issues. At training time, TensorFlow does require a fixed tensor size, but you can...

I think TF2 no longer supports tf.contrib (https://www.tensorflow.org/guide/migrate) I guess you need to find a way to replace the two tf.contrib.layers used: tf.contrib.layers.xavier_initializer() tf.contrib.layers.batch_norm

Hi Cindy, I've added a download link for the single and multi-view network for the 3dnin fc architecture at the end of the Results section in README. Hope it helps.