Open hanslovsky opened 5 years ago
I slightly misunderstood the group and dataset hierarchy inside h5 files in BDV format:
Instead of
<time>/<setup>/<scale>/<name>`
It actually is
<time>/<setup??>/<scale??>/cells
Where cells
is always the same name. This makes importing and discovering single time points of single setups a lot more compatible multi-scale support in Paintera.
Thanks to @tpietzsch for clarifying!
What we should do (in my opinion) is
AffineTransform3D
).This is much simpler than fully supporting BDV format. I will leave this issue open because it might be good to actually have full support of BDV format at some point.
While based on BDV, Paintera currenlty does not support the BDV data format. @tpietzsch suggested we add least add some rudimentary support for it so people in the SPIM world have access to Paintera.
Quoting myself from an e-mail that I sent to @tpietzsch
I had a look at the structure of the BDV example h5 (drosophila.h5) file at http://fly.mpi-cbg.de/~pietzsch/bdv-example/ If I understand correctly, the actual dataset is not located at
but rather in
which, if I parse correctly, results in a scheme
which is not trivially compatible with Paintera's currently supported (but discouraged in favor of Paintera's own format) multi-scale format
The difference is, as far as I understand, that Paintera treats the