search

tsunghan-wu / ReDAL

🍀 Official pytorch implementation of "ReDAL: Region-based and Diversity-aware Active Learning for Point Cloud Semantic Segmentation. Wu et al. ICCV 2021."
41 stars 10 forks source link

Image segmentation #4

Closed abhigoku10 closed 2 years ago

abhigoku10 commented 2 years ago

@tsunghan-mama hi thanks for sharing this wonderful code base i had few queries

  1. Does this source code support or can be user for point cloud data lidar for automotive driving scenario
  2. Does this source code also support 2d image segmentation Thanks in advance
tsunghan-wu commented 2 years ago

Hi,

  1. Yes. This code is applicable to self-driving scenario as well as indoor robotics. Specifically, the code supports SemanticKITTI dataset (outdoor scene) and S3DIS dataset (indoor room).
  2. Yes. This method can also be used on 2D image segmentation. In fact, we did further experiments on the 2D dataset few month ago and found that region-based selection highly reduces the annotation burden. (You can refer to thin link to see how the approach can be used in 2D.)

Hope these information help.

Best, Tsung-Han

tsunghan-wu commented 2 years ago

Close the issue due to inactivity for a long time.