OpenGVLab / PonderV2

PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm

https://arxiv.org/abs/2310.08586

OpenGVLab/PonderV2 Issues

DTU image synthesis experiment
Updated a month ago1
Visualization of RGB&Depth images
Closed a month ago1
Query about the preprocess of ScanNetpp
Updated 2 months ago4
Will NuScenes dataset be supported?
Closed 3 months ago
Can you provide per-class IoU for semantic segmentation on S3DIS 6-fold?
Closed 3 months ago1
About the code for Image Synthesis From Point Clouds
Closed 3 months ago1
3D Object Detection Code
Closed 3 months ago1
Question about experiment folder created by training script.
Closed 3 months ago2
Visualize Code
Closed 4 months ago3
Feature extraction for point clouds
Closed 4 months ago1
Share ScanNet skip.lst
Closed 4 months ago3
Change method for rendering decoder
Closed 4 months ago1
Readme for Instance Segmentation
Closed 4 months ago8
how to use the network for semantic segmentation?
Closed 5 months ago1
Training w/ default config diverges and ultimately crashes: NaN or Inf found in input tensor
Closed 6 months ago2
请问室内场景在训练的时候，如果没有 rgb 信息，同样需要做颜色的渲染监督吗？在训练的时候遇到没有 rgb 信息的点云时候是怎么做的呢？训练的用于渲染的 shallow MLP 是所有场景共享的，还是每个场景有各自的 MLP 参数？
Closed 8 months ago1
Will this project support KITTI dateset in the future? : )
Closed 8 months ago1
How can I run the downsteam tasks with your model?
Closed 8 months ago2
Estimated release date
Closed 8 months ago3
Some question about outdoor rendering results
Closed 8 months ago4
Looking forward to seeing more instructions and code
Closed 8 months ago3