IIT-PAVIS / Look_Around_And_Learn

Code for the paper titled "Look Around 👀 and Learn 🎓: Self-Training Object Detection by Exploration" accepted at ECCV2024
MIT License
9 stars 0 forks source link

pseudo labeler with semantic mapping and softmax prediction vector #7

Open yuanxuanS opened 1 day ago

yuanxuanS commented 1 day ago

I found no function of compute soft softmax pseudo labels and project masks from 3D pointcloud to images. Could you please tell me which consensus labeler did you use?

morpheus1820 commented 1 day ago

The projection of pseudo labels 3D pcd to 2D masks in images is done by this method: https://github.com/IIT-PAVIS/Look_Around_And_Learn/blob/e939597aac5de4c8a7f173bbf728cd2dc98d5ec8/experimenting_env/detector/pseudolabeler.py#L200