Closed JiyunJang-24 closed 7 months ago
Hello, there is the 20% kitti info. Also, are you loading parameters from our pre-trained checkpoint correctly? The BACKBONE_3D of our pre-training model is "VoxelResBackBone8x", so it should also be aligned in the config of SECOND-IOU and SECOND.
Thank you for answering. I was using voxelbackbone8x as 3d backbone while using second. Now that I'm using voxelResbackbone8x, I'm seeing performance improvements like paper. Thank you!
I would like to ask a question about the AD-PT paper. In the AD-PT paper, when using only 20% of the kitti dataset, it showed a significant performance improvement in the SECOND model. However, when I used the SECOND-IOU model using the ONCE 1M provided as a backbone pretrain model, it showed -1.7mAP performance when using AD-PT. I would like to ask if you can release the imageset of the data when using 20% of the kitti.