TAO MTL support training for detection and regression

For multitask learning with TAO there is a notebook for multi task classification. Does TAO MTL support for object detection and regression as well ?
for example is it possible to perfom MTL on face detection + facial landmark detection + gender classification …etc

There is only multi task for classification. For your case, it is similar to deepstream_tao_apps/apps/tao_others/deepstream-emotion-app at master · NVIDIA-AI-IOT/deepstream_tao_apps · GitHub. We can leverage the pipeline.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.