yjcaimeow

Train 2D feature extraction and point representation

6

Hi @Xharlie and @zexxu , I do not understand about "point feature initial" part. Does the model under "init" folder for all dataset? Or for different dataset, I need to...

batch size can not bigger than 1 !?

1

Dear Author, It seems the code does not supper batch_size >1? Because the code here? https://github.com/Xharlie/pointnerf/blob/master/models/neural_points/query_point_indices_worldcoords.py#:~:text=R%20%3D%20torch.max(torch.sum(ray_mask_tensor.to(torch.int32))).cpu().numpy() Bests

HI @YANG-H, Could you please provide the metric evaluation code for real data for Consistency, Fidelity and MMD? Best, YIngjie CAI

loss value

4

Dear author, I retrained the shapenet model with spare point cloud as input. The initial loss values are about 20000. Is it normal since I think the loss is really...

 I can not understand why transform obj to off form. Why not just change the line 33th from "mesh = trimesh.load(path + '/isosurf.off', process=False)" to "mesh = trimesh.load(path +...

train_data.h5 filenames

2

Hi, could you please provide the name list of train data? @xiaogangw Thanks

 Dear @Xharlie I use the offical code on scannet dataset. The training process is unstable. Did you meet the problem? Best Yingjie CAI

How pixel assign to cell?

1

Dear @jaharkes @hturki @therealsatya @teiszler Thanks the great work! But I have some understing. How to assign a pixel to certain 3D cell? Using the depth of the pixel? There...

How do I know when to stop Stage 2 training?

3

Hi @ZENGXH , Is there any indicator to judge whether stage2 is trained enough? I used a different data set, I don't know what epoch/iteration to stop. Best regards, Yingjie