search

yinjunbo / ProposalContrast

This repository contains the PyTorch implementation of the ECCV'2022 paper, ProposalContrast: Unsupervised Pre-training for LiDAR-based 3D Object Detection.
MIT License
53 stars 4 forks source link

How do you re-implement PointContrast and DepthContrast? #8

Closed Hiusam closed 1 year ago

Hiusam commented 1 year ago

Hi, thank you for your excellent work.

I wonder how you re-implement PointContrast and DephtContrast in your codebase? Can you give me some hints?

If you can provide the code (or some code snippets), I will appreciate it very much.

yinjunbo commented 1 year ago

Hi, thanks for you interest. We implemented PointContrast by randomly sampling points from two augmented views and then enforcing point-wise contrastive loss. This can be easily realized by our codebase. As for DephtContrast, we directly followed their offcial code https://github.com/facebookresearch/DepthContrast.