smileyenot983

I have printed angle to scanID correspondings and for me it is like that: angle -14.999978 scanID 0 angle -12.999980 scanID 1 angle -10.999983 scanID 2 angle -8.999986 scanID 3...

It is VLP-16, just like you

i fixed it by changing following lines in CMakeLists.txt: lines 132-150 into: ``` #-------------------- # DEPENDENCY: eigen 3 #-------------------- # find_path(EIGEN_INCLUDE_DIR NAMES signature_of_eigen3_matrix_library # HINTS ENV EIGEN3_INC_DIR # ENV EIGEN3_DIR...

After i build it for the second time i had same error, i have found discussions about creating custom ROS messages and found difference between their CMakeLists and this repo's....

Was someone able to solve this? I am having same error on 4090

> Can you please check if any other process is running on your GPU? Here is the output of nvidia-smi: ``` +---------------------------------------------------------------------------------------+ | NVIDIA-SMI 535.183.01 Driver Version: 535.183.01 CUDA Version:...

Ok, looks like everything got fixed after googling. I was able to solve it by updating the pytorch and torchvision to 2.1.2: `pip install torch==2.1.2+cu118 torchvision==0.16.2+cu118 --extra-index-url https://download.pytorch.org/whl/cu118` Hope this...

Any updates on this? I found out that in arkitscenes preprocessing not all data from `raw` set is used

Is there any update? I am using the pretrained weights too, but i got only around 20 mIoU. Wondering what is the reason

Upd. I found out that same conv network can be applied to 3d points as the input dimensionality(by default=16) is same. Issue might be closed