exiawsh
Incorporating Historical Frame Points in StreamPETR
I am currently working on extending the LiDAR modality within the StreamPETR framework. However, I am unsure how to incorporate historical frame sweep points into the data{} during the forward_train process.
From my understanding, it seems that I need to modify the nuscenes_converter.py file and regenerate the PKL files. Could someone provide guidance on how to achieve this?
Also I'm confused about the variable names here, is the gradient being calculated here using the loss of two frames? https://github.com/exiawsh/StreamPETR/blob/95f64702306ccdb7a78889578b2a55b5deb35b2a/projects/mmdet3d_plugin/models/detectors/petr3d.py#L104-L137 https://github.com/exiawsh/StreamPETR/blob/95f64702306ccdb7a78889578b2a55b5deb35b2a/projects/mmdet3d_plugin/models/detectors/petr3d.py#L264-L271
https://github.com/exiawsh/StreamPETR/issues/232#issuecomment-2408789991