[PyTorch] Official implementation of CVPR2022 paper "TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection with Transformers". https://arxiv.org/abs/2203.11496
The paper explains about implementation details that the weights of the image backbone network was frozen during training.
So I wonder how the performance would be if the weights of the image backbone network was not frozen during training.
have you ever tried to train the model without freezing the image backbone network?
The paper explains about implementation details that the weights of the image backbone network was frozen during training. So I wonder how the performance would be if the weights of the image backbone network was not frozen during training. have you ever tried to train the model without freezing the image backbone network?