raphaelsulzer

raphaelsulzer

Member Since 5 years ago

Experience Points
4
follower
Lessons Completed
15
follow
Lessons Completed
47
stars
Best Reply Awards
31
repos

237 contributions in the last year

Pinned
⚡ A python implementation to extract Shape DNA [1] of triangle surface meshes
⚡ A serious game for spatial planning in the form of a touch-supportive plugin for QGIS
⚡ A QGIS plugin to interactively label vector layer features
⚡ the LASTIG homepage of Raphael Sulzer
Activity
Oct
22
1 day ago
Oct
21
2 days ago
Oct
20
3 days ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

added lots of different options for point sampling on the lines of sight

commit sha: b01473b47b28d040f3072667af182f8af646ac13

push time in 3 days ago
Oct
12
1 week ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

fixed generation and evaluation of models with sensors.

commit sha: e25af131abdbb88e2b723ee3ec26552821e07c76

push time in 1 week ago
Oct
8
2 weeks ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

started an out-of-category experiment on modelnet10

commit sha: 521b2111c79071050a72b4cfe1abd1d3fcd24e89

push time in 2 weeks ago
Oct
5
2 weeks ago
Activity icon
issue

raphaelsulzer issue comment autonomousvision/occupancy_networks

raphaelsulzer
raphaelsulzer

What can I do if I only want to use the MISE part?

Hello, now I have some 3D object points(in .npy file) and I want to use the method MISE described in your paper to construct a mesh . But how can I do that?

raphaelsulzer
raphaelsulzer

Hi, could you explain the 'threshold' parameter of the MISE code?

Oct
1
3 weeks ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

added a mygeneration.py script for using the MISE function

commit sha: 40729c32a23e3be085f69d533274320a152caed5

push time in 3 weeks ago
Sep
17
1 month ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

added option to add points along the sensor to the pointnet

commit sha: f6f5998a2aa77fa9f291a8d153cc8779617102c7

push time in 1 month ago
Activity icon
issue

raphaelsulzer issue autonomousvision/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

Watertight ShapeNet models

Hi,

would it be possible to share the watertight models of the ShapeNet subset you used?

Kind regards!

Sep
16
1 month ago
Activity icon
fork

raphaelsulzer forked tensorflow/graphics

⚡ TensorFlow Graphics: Differentiable Graphics Layers for TensorFlow
raphaelsulzer Apache License 2.0 Updated
fork time in 1 month ago
started
started time in 1 month ago
Sep
15
1 month ago
Sep
10
1 month ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

complete pipeline working with added sensor position

commit sha: fa15bad0e0fda5d155b6b0fd529efcf000d2bb11

push time in 1 month ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

complete pipeline working with added sensor position

commit sha: 7e61febc701adbb0458edcedb4c20d7c175ae601

push time in 1 month ago
Sep
9
1 month ago
push

raphaelsulzer push raphaelsulzer/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

remove a bug for sliding window

raphaelsulzer
raphaelsulzer
raphaelsulzer
raphaelsulzer

Merge branch 'master' of github.com:autonomousvision/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

update torch-scatter installation

raphaelsulzer
raphaelsulzer

Merge github.com:autonomousvision/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

commit sha: 8283dd0797748cad8f7fd3b83ec8936502f4c8be

push time in 1 month ago
Jul
28
2 months ago
Activity icon
issue

raphaelsulzer issue autonomousvision/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

No noise in input point clouds?

Hi,

thank you very much for making this great work available here.

I was experimenting with retraining your network for which I downloaded the preprocessed ShapeNet and synthetic room dataset. You state in the paper that you "sample 3000 points from the mesh and apply Gaussian noise with zero mean and standard deviation 0.05" to generate noisy input point clouds. (I understand the reasoning in this issue, and can see in the config files that it is actually 0.005.)

However, to me it seems like there is no added noise in neither the pointcloud.ply nor pointcloud.npz files in both the ShapeNet and synthetic room dataset. If I follow your approach correctly these files are generated with the sample_mesh.ply script from the occupancy networks repository. In the export_points function for generating occupancy samples noise is added, however not in the export_pointcloud function, which is used as reconstruction input. Shouldn't this be the case, since you postulate to reconstruct meshes from noisy point clouds?

Kind regards

Activity icon
issue

raphaelsulzer issue comment autonomousvision/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

No noise in input point clouds?

Hi,

thank you very much for making this great work available here.

I was experimenting with retraining your network for which I downloaded the preprocessed ShapeNet and synthetic room dataset. You state in the paper that you "sample 3000 points from the mesh and apply Gaussian noise with zero mean and standard deviation 0.05" to generate noisy input point clouds. (I understand the reasoning in this issue, and can see in the config files that it is actually 0.005.)

However, to me it seems like there is no added noise in neither the pointcloud.ply nor pointcloud.npz files in both the ShapeNet and synthetic room dataset. If I follow your approach correctly these files are generated with the sample_mesh.ply script from the occupancy networks repository. In the export_points function for generating occupancy samples noise is added, however not in the export_pointcloud function, which is used as reconstruction input. Shouldn't this be the case, since you postulate to reconstruct meshes from noisy point clouds?

Kind regards

raphaelsulzer
raphaelsulzer

Yes, that makes it clear. Thank you!

Jul
26
2 months ago
Activity icon
issue

raphaelsulzer issue autonomousvision/convolutional_occupancy_networks

raphaelsulzer
raphaelsulzer

Noise added to point clouds

Hi,

thank you very much for making this great work available here.

I was experimenting with retraining your network for which I downloaded the preprocessed ShapeNet and synthetic room dataset. You state in the paper that you "sample 3000 points from the mesh and apply Gaussian noise with zero mean and standard deviation 0.05" to generate noisy input point clouds. (I understand the reasoning in this issue, and can see in the config files that it is actually 0.005.)

However, to me it seems like there is no added noise in neither the pointcloud.ply nor pointcloud.npz files in both the ShapeNet and synthetic room dataset. If I follow your approach correctly these files are generated with the sample_mesh.ply script from the occupancy networks repository. In the export_points function for generating occupancy samples noise is added, however not in the export_pointcloud function, which is used as reconstruction input. Shouldn't this be the case, since you postulate to be able to reconstruct meshes from noisy point clouds?

Kind regards