Open Images Pre-trained DetectNet_v2
====================================

.. _oi_pt_dnv2_intro:

Object detection is a popular computer vision technique that can detect one or multiple objects in a frame. Object detection will recognize
the individual objects in an image and place bounding boxes around the object. This model object contains pretrained weights that may be
used to initialize the **DetectNet_v2** object detection networks in Transfer Learning Toolkit (TLT) to facilitate transfer learning.

It is trained on a subset of the Google OpenImages dataset. 

Supported Backbones
-------------------

The following backbones are supported with DetectNet_v2 networks:

- resnet10/resnet18/resnet34/resnet50
- vgg16/vgg19
- googlenet
- mobilenet_v1/mobilenet_v2
- squeezenet
- darknet19/darknet53

.. Note::
     - These are unpruned models with just the feature extractor weights, and may not be used without re-training in an object detection application
     - When using the ResNet34 model, please set the :code:`all_projections` field in the :code:`model_config` to :code:`False`. For more information about this parameter,
       please refer to the :ref:`model config section in the detectnet_v2 documentation <model_config_detectnet_v2>`.
     - The pre-trained weights in this model are only for DetectNet_v2 object detection networks and shouldn't be used for
       YOLOV3/YOLOV4, RetinaNet, FasterRCNN, SSD and DSSD based object detection models.


You may find more information about these models on NGC_.

.. _NGC: https://ngc.nvidia.com/catalog/models/nvidia:tlt_pretrained_detectnet_v2