Closed ShashankBice closed 2 years ago
Using --save-tif-image
saves the day, although "new" tif image has to be saved, and looks like the dimensions are cropped by a few pixels on the edges. Will continue investigating, and do full run from rpcs obtained by updated frame camera models. To me it looks like that although I still provide updated rpc xml, the code still reads the old, original RPC values in the input tif image or something.
Closing this as --save-tif-image
works well.
Work with Oleg to nail down the best parameters for exporting RPC model from the improved, self-consistent and geo-registered frame cameras.