Rana Hanocka

Hi, All the convolutional kernels have the same size (5). There are 4 values applied for the 1 ring neighbors, and 1 value applied on the edge itself. In the...

Hi @MaroonAmor , Arc cosine is defined between -1 ≤ x ≤1, and so due to any floating point arithmetic, we ensure this is the case with the clip, here:...

Hi @MaroonAmor , I think that signed angles will be sensitive to order. Since we don't have a way to consistently define which face comes first, this just tells us...

Hi @yuvaramsingh94 , I have not personally performed manual segmentation (by hand) to create a dataset, so I am not sure. It seems that some people have done that with...

Hi @duhau , Check [this answer](https://github.com/ranahanocka/MeshCNN/issues/54#issuecomment-569809104), you basically need to export the .edges from the mesh class. You can just create the seseg file by looking at the segmentation of...

Hi @theveloped , This repo is doing supervised classification, so it requires a training set of your desired classes (e.g., nuts and bolts). If you don't have such a dataset,...

Hi @shangguan91 , I guess there are a lot. You can check the references within our paper (who we compared against for the mesh classification, etc). You can also check...

Hi @manuelknott , There is a viewer script included in this repo (e.g.,` bash ./scripts/human_seg/view.sh` ). These days I am using [polyscope](https://github.com/nmwsharp/polyscope), which is fast and can plot per-edge colors....

> (1.1)Do you have experience with "sticked-together", but not properly connected geometry? The decimation process seems to mess up the meshes really bad in case of low resolution parts and...

Hi @liangqianqian123 , it means number of input channels (input features). In the case of this work, opt.input_nc = 5 (always). It is the 5-dimensional geometric feature vector. It is...