Detectron: Only objects described in terms of polygons are included in training
Hi Detectron,
According to json_dataset.py, it seems that all crowd objects are excluded from training:
Therefore all objects (used in training) are described in terms of polygons, i.e., one object, one polygon. All crowd objects (i.e., one object, more than one polygons) will be ignored. Am I correct?
Thanks!
realwecan
All 3 comments
It seems that you are correct, maybe it's for saving memory in the training process.
ambigus9
on 15 Apr 2018
C bc h.fig
Zoalfekar536
on 15 Apr 2018
Crowd "objects" are cases where there are too many object instances to accurately annotate. E.g., imagine a sports stadium with hundreds of tiny people in the seats -- the whole mass of people will be marked as a crowd. These should be excluded from training because they are not individual object instances.
rbgirshick
on 28 Apr 2018
👍3
Was this page helpful?
0 / 5 - 0 ratings
Related issues
pacelu
·
3Comments
fangpengcheng95
·
4Comments
kleingeo
·
3Comments
coldgemini
·
3Comments
rbgirshick
·
3Comments
Most helpful comment
Crowd "objects" are cases where there are too many object instances to accurately annotate. E.g., imagine a sports stadium with hundreds of tiny people in the seats -- the whole mass of people will be marked as a crowd. These should be excluded from training because they are not individual object instances.