Shirokuma (k tanaka)

I have not yet figured out what we should do about handling flann. I also need to understand vcpkg better. I don't foresee any problems with migrating common libraries such...

Hi @baoyufuyou , Thanks for the feedback. The collision module has some APIs available, but it has not been fully tested yet. I plan to add more python examples in...

I added some python examples for collision detection. https://github.com/neka-nat/cupoch/blob/master/examples/python/robotics/collision_linset_voxel.py https://github.com/neka-nat/cupoch/blob/master/examples/python/robotics/collision_voxelgrid.py

Hi @Apeiria , Thank you for reporting. ScalableTSDF uses the unordered_map implemented in STDGPU, but it is not very memory efficient. I am currently exploring ways to use memory more...

Hi, Your suggestion is correct. The grid_index can be eliminated and the size of the TSDFVoxel can be further reduced. However, I find the problem is that scalable tsdf uses...

Hi @VladMVLX , Thank you for your feedback. The implementation of scarable tsdf will be done in the future. It need to be modified for CUDA based on the open3d...

I implemented TSDF generation and point cloud extraction The implementation of mesh extraction is not yet done. Also, as you can see in #12 , it's not memory efficient.

Hi @dujinyu , Thank you for your feedback. The slow computation and drawing discontinuity is due to the computational load of autograd. As for the algorithm, it will be renamed...

HI @TianrongChen, Thank you for your reporting. If possible, can you push the changes to your forked repository? I'll clone it from there and examine it.

Hi @NikEyX, Thank you for your trying! The implementation of R2D2 is made experimentally. I couldn't launch 256 actors like in the paper, so I tuned some parameters but couldn't...