xuelin-chen / pcl2pcl-gan-pub Goto Github PK
View Code? Open in Web Editor NEWUnpaired Point Cloud Completion on Real Scans using Adversarial Training
Unpaired Point Cloud Completion on Real Scans using Adversarial Training
Could you please explain why you need to manually align these objects from Scannet to be consistently orientated with models in ShapeNet dataset?
Dear author:
Would you please provide the pretrained models?
I found your gmail in your github page and sent emails to you,but it seems that your gmail address is invalid.
Hi,The dataset download method you provided has invalid link. How can I download your dataset?thanks a lot
Thanks for your excellent work.
I want to use some real scanned data as input, Could you explain how to align real scanned data with the shapenet data, which software or algorithm could help?
Thanks!
Hi, xuelin,
I am really interested in your work, and can you provide or give some suggestions about how your extract the partial object from Scannet?
Thanks!
Because the data processing seems memory-consuming, can you tell me the memory used to process the data.
Thanks for your excellent work.
Could you please share the code for extracting and aligning objects for scannet dataset?
Hi Chen,
The work in this paper is quite amazing seeing the results.
Can you shed some light on how you use the EMD and the Hausdorff distance to calculate the loss for the point clouds? Can you share some code for it that can be plugged in right away?
Actually, we are implementing some network that can be greatly helped by your implemention.
If you can share the details and share the code for the EMD and Hausdorff distance, it would be of immense help to us.
Thanks,
Prashant
Hi, Chen~
Sorry to bother you. I have a problem when I run train_pcl2pcl_gan_3D-EPN.py.
train GAN: python train_pcl2pcl_gan_3D-EPN.py
After the training, I use Meshlab to see the result in pc2pc/run_3D-EPN/run_car/pcl2pcl/log_car_pcl2pcl_gan_3D-EPN_default_hausdorff/fake_cleans. However, the ply point clouds in file(reconstr_x) are all like wool ball. GAN didn't seem to work.
I don't know what the problem is. Maybe the dataset.
Here is what I did:
I use the dataset "shape_net_core_uniform_samples_2048" (from other projects) . And I use matlab to make incomplete point set.
And use pc2pc/data_processing to make pickle file.
python train_ae_ShapeNet-v1.py
python train_ae_3D-EPN.py
(the ply point clouds (ShapeNet-v1 and 3D-EPN) in reconstr are successful.)
But after I run "python train_pcl2pcl_gan_3D-EPN.py", the ply clouds in reconstr are a mess.
I also reduce the number of batch_size to run it. Does it affect the outcome?
If I don't clarify my question, tell me what I should show.
I'm a novice. I really hope to get your help.
Hello! Is there a test part of your code?
It seems like that the link for downloading the data points to a missing page.
Could you please provide an alternative way for download or fix that link.
Thanks!
Hello, I want to train your network on KITTI Dataset, but I don't know how to get the data of ground truth of KITTI Dataset and the Complete 3D point cloud of cars, can you tell me how to get these data?
Hi, Thank you for open sourcing the code. It's a good work.
When running the code, I went into a problem. I ran CUDA_VISIBLE_DEVICES=0 python3 train_ae_ShapeNet-v1.py and got an error.
Traceback (most recent call last):
File "train_ae_ShapeNet-v1.py", line 104, in
TRAIN_DATASET = shapenet_pc_dataset.ShapeNetPartPointsDataset_V1(para_config['point_cloud_dir'], batch_size=para_config['batch_size'], npoint=para_config['point_cloud_shape'][0], shuffle=True, split='trainval', preprocess=False)
File "/home/user05/Shuyan/pcl2pcl-gan-pub/pc2pc/shapenet_pc_dataset.py", line 430, in init
self.point_clouds = self._read_all_pointclouds(self.point_cloud_dir)
File "/home/user05/Shuyan/pcl2pcl-gan-pub/pc2pc/shapenet_pc_dataset.py", line 477, in _read_all_pointclouds
rotated_points = pc_util.rotate_point_cloud_by_axis_angle(pc, [0,1,0], 90)
File "/home/user05/Shuyan/pcl2pcl-gan-pub/utils/pc_util.py", line 564, in rotate_point_cloud_by_axis_angle
rot_m = pymesh.Quaternion.fromAxisAngle(axis, angle)
AttributeError: module 'pymesh' has no attribute 'Quaternion'
I want to know which version of pymesh you are using, or is your pymesh downloaded from GitHub?
Thanks for your great works!
The download link for data is not working now, could you please update it ?
Thank you for your great work! can you provide your visualization scripts to render the point cloud?
Hi, could you provide your pre-trained models?
There are 10 files in the dataset you privided. I download all the files and try unzip command, but I got an error message:
error: invalid zip file with overlapped components (possible zip bomb)
How to solve this problem? The platform I used is Ubuntu 18.04 LTS.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.