This is the code related to "Sparse-to-dense Feature Matching: Intra and Inter domain Cross-modal Learning in Domain Adaptation for 3D Semantic Segmentation" (ICCV 2021).
hello author, thanks to your efforts and code. Recently, I implement your code in 12G 3060 GPU with you the same environment in pytorch=1.7.1, cuda=11.1, python=3.7. I find 2d+3d iou only has 59.39 which is gap with your article results. @leolyj@DuoPeng-CVer
hello, author. Recently, I have trained the model, but its' performance has great gaps whith your results in your paper. At the same time, why code has great difference with idea in paper? @leolyj@DuoPeng-CVer