Hi, I succeeded to train and evaluate the model (maybe it is SECOND-v1.0).
Now I want to know how to do the feature fusion (feature alignment and aggregation) process.
Anyone can help this?
More specifically, about the given feature files (in model_sec_car_37144/predict_test/step_37144/log/)
there are 2 voxel features (#20, #75) and 3 spatial features (#75, #98, #99)
Are these about KITTI data or T&J data? and are these already aligned?
Hi, I succeeded to train and evaluate the model (maybe it is SECOND-v1.0). Now I want to know how to do the feature fusion (feature alignment and aggregation) process. Anyone can help this?
More specifically, about the given feature files (in model_sec_car_37144/predict_test/step_37144/log/) there are 2 voxel features (#20, #75) and 3 spatial features (#75, #98, #99) Are these about KITTI data or T&J data? and are these already aligned?