I’ve been trying to do multi object detection, and I got it working using ObjectItemList. However, I struggled for a long time, and I feel like the docs are misleading, so I’d like some clarification. Here’s the documentation on these two.
class
ObjectCategoryList
[source][test]
ObjectCategoryList
(items
:Iterator
[T_co
],classes
:Collection
[T_co
]=None
,label_delim
:str
=None
,one_hot
:bool
=False
, ****kwargs
** ) ::MultiCategoryList
ItemList
for labelled bounding boxes.
class
ObjectItemList
[source][test]
ObjectItemList
( ***args
** ,convert_mode
='RGB'
,after_open
:Callable
=None
, ****kwargs
** ) ::ImageList
ItemList
suitable for object detection.
First question. is a “labelled bounding box” not the same as “object detection”?
Second, how does ObjectCategoryList expect its class input? I was trying to use from_df
or from_csv
and didn’t have much luck. I tried to walk through the debugger, but I think my understanding of what the label is, or how it should be structured.
Third, not a question, but a statement. There are zero tests for ObjectCategoryList. I think it would be helpful to have a test to understand how the input should look. I’d love to help, if the first two questions can be answered.