Your browser was unable to load all of the resources. They may have been blocked by your firewall, proxy or browser configuration.
Press Ctrl+F5 or Ctrl+Shift+R to have your browser try again.

complex models for object detection? #2

Hello,

Any guide or pointers to set up a model definition for tasks such as object detection which take in images, bounding boxes as input features and output bounding boxes of the objects detected in the image?

  • solved #2
  • replies 2
  • views 3.1K
  • likes 1
#2

Hi Pavan,
We don't support that yet, we were thinking about adding an additional bounding box datatype with an output decoder, but it's not quite there yet. There are a bunch a specific projects that do only that, so if that's your use case you have plenty of options.
You can really vaguely approximate it with a certai number of numeric outputs in Ludwig (for instance 4 if you are looking for just one bounding box), but as there is no specific decoder I doubt it's going to work well.
Check out the next releases of Ludwig to see when we'll add it!

pavan4 accepted post #2 as the answer
#3