You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently annotations for object-detection of images can have box or polygonal regions. Add bit-mask regions as well, so conversions between formats that use this type of region internally (e.g. TFRecords) won't have to enclose them with a polygon and then re-raster, which can introduce dissimilarities.
The text was updated successfully, but these errors were encountered:
Currently annotations for object-detection of images can have box or polygonal regions. Add bit-mask regions as well, so conversions between formats that use this type of region internally (e.g. TFRecords) won't have to enclose them with a polygon and then re-raster, which can introduce dissimilarities.
The text was updated successfully, but these errors were encountered: