UT-Austin-RPL / GIGA

Official PyTorch implementation of Synergies Between Affordance and Geometry: 6-DoF Grasp Detection via Implicit Representations
MIT License
131 stars 26 forks source link

visualize grasp afforance #11

Closed ccc1711 closed 8 months ago

ccc1711 commented 2 years ago

May I ask what software did you use to visualize the grasp afforance? colored like this。。。 1647106843(1)

Steve-Tod commented 2 years ago

Hi, you can refer to this code snippets: https://github.com/UT-Austin-RPL/GIGA/blob/d67c4388d334babe6c11c1555a6e848fb4828c84/src/vgn/utils/visual.py#L101

ccc1711 commented 2 years ago

ok,and how can I get the size of feature map for each layer?

Steve-Tod commented 2 years ago

I think you can print the size of the different layers of the network and compute the size of the feature map.

ccc1711 commented 2 years ago

Is the “TSDF-3DCNN-PointNet-3DUnet” process reasonable?

Steve-Tod commented 2 years ago

Sorry for the late reply, I didn't see this earlier. What do you mean by “TSDF-3DCNN-PointNet-3DUnet”?

ccc1711 commented 2 years ago

It's nothing ,i was wrong, and is the process of projection and aggregation implemented in def generate_frid_features()?

Steve-Tod commented 2 years ago

Yeah, that's right.