Query on the Learned_Representation

[TTiannaiTT]

Hello! I got a really confused result when using this code:

Program mIOU
gen1_optimized_augment 0.5425813411258469
gen1_EST1 0.0010877805692414868
gen1_EST2 0.0010434959245672268

It seems like that the learned representation have a really poor performance for detection on Gen1.
I have also tested the ERGO-12 but it is aligned with the paper.
Do you have any idea about this problem? Thanks in advance.

[nikky4D]

@TTiannaiTT Were you able to solve this issue with the poor performance on GEn1? if so what did you do to the input representation?