Global Game Jam 2024

Wow, what a weekend. The Global Game Jam is intense. But I and 3 other members manage to build a game for the theme "Make me laugh".

Go and check out the Apocalyptic Bowling! Destroy everything!

Rigid Registration Finished

Yay! Finally! After 3.5 weeks of work, it is done. And it even has an animated GIF to undertand how it works through the GUI. 

I'll just try to install the code all over again to build a step-by-step tutorial so anyone can use it.

Check out the code here:

GitHub/LilianeRA/RigidRegistration

Visualization almost complete

Another update for the Rigid Registration project.

Now you can visualize the result of the registration, choose the method configuration from the GUI and reset to check the performance of another method.

There are still some issues to investigate, but the first release is close.

See you soon!

Lie Groups Incorporated

More news. Today was really productive. 

I've just finished the implementation of all the Lie Group approaches into the ICP Rigid Registration method, according to my paper:

Pairwise Rigid Registration Based on Riemannian Geometry and Lie Structures of Orientation Tensors

I still need to confirm if the results are the same as in the original code.

But you can still check the code here:

GitHub/LilianeRA/RigidRegistration

Basic Rigid Registration complete

Hello there.

I've finished the base code for the Original ICP, the ICP-CTSF, SWC-ICP and SWC-CTSF. I've only tested it with the Stanford Bunny, downsampled to have 894 points. After incorporating the Lie Groups, I'll start the real performance test.

Have a look!

GitHub/LilianeRA/RigidRegistration

I'm back. Again

Sorry guys, I can only come back here during my vacations. But this time, I made a new year's promise to give more attention to this blog. Right now, I'm developing a 3D Visualizer based on OpenGL and Dear ImGui for me to use in any other project. Details here:

GitHub/LilianeRA/Visualizer3D

And the second project is the code refactor for the official implementation of the Rigid Registration methods developed by my college colleagues: ICP-CTSF and SWC-ICP. The code is being published here:

GitHub/LilianeRA/RigidRegistration

For now, I've only tested it on Windows 10. Feel free to try to install it. If you have any trouble, you can let me know

Tensorflow 1.x to 2.x

I changed the only function from Tensorflow 1.x of TempoGAN that I mentioned in this post, from

tf.contrib.layers.batch_norm()

to

tf.compat.v1.layers.batch_normalization()

and, of course, the pre-trained weights are useless now that I changed the net structure. So, I need to retrain it, but there is no script to do it. I can do one myself and be very careful when choosing the parameters. Or, I use Google Colab to run the original net for me.

You may have noticed that Google Colab does not support Tensorflow 1.x anymore, right? But no worries. I've found a way to overcome this with miniconda here.