DeepIM-PyTorch
DeepIM-PyTorch copied to clipboard
NVlabs
PyTorch implementation of the DeepIM framework
I notice that you make a real-world demo that demonstrates estimating the pose of real ycb-v objects. I cannot find where to buy these this object. Can you give some...
``` ================================================= loading 3D models libEGL warning: failed to open /dev/dri/renderD131: Permission denied libEGL warning: failed to open /dev/dri/renderD131: Permission denied Unable to initialize EGL ``` I was running ./experiments/scripts/demo.sh...
the requirements say it needs Ubuntu 16.04 or above. does it run on windows too? if not any idea to make it run. thanks in advance
## Short description On Linux, DeepIM-PyTorch can't run `./experiments/scripts/demo.sh`. The only log/error message are: ``` loading 3D models libEGL warning: DRI2: failed to create dri screen libEGL warning: Not allowed...
I'm trying to build the project, I'm running ubuntu 20.04, PyTorch 1.8.0 and Cuda 11.4. At the fourth bullet point ``` cd $ROOT/lib/point_matching_loss sudo python setup.py install ``` I get...
Hello! First of all, thank you for making this code available, it look awesome! I have a custom dataset (including object models, background images, and region of possible camera poses...
I'm trying to execute the demo.sh and I'm getting errors at line 364 of ycb_renderer.py when loading material properties. It's expecting an array when the property is an int. Here...
As titled, will the support for the LM-O dataset be added? Or how should I modify the code to make ycb_renderer work on Linemod objects? Thanks!
The download links in the readme is not accessible. Please provide alternative links. No links in [this](https://github.com/NVlabs/DeepIM-PyTorch#training-and-testing-on-the-ycb-video-dataset) section works now. Please to help
FILE: ycb_video.py (in lib>datasets) LINE ~ 77,78 # 3D model paths self.model_mesh_paths = ['{}/{}/textured_simple.obj'.format(self._model_path, cls) for cls in self._classes_all] self.model_texture_paths = ['{}/{}/texture_map.png'.format(self._model_path, cls) for cls in self._classes_all] This is asking...
