NGC | Catalog
CatalogModelsPre-trained Deformable DETR ImageNet weights

Pre-trained Deformable DETR ImageNet weights

For downloads and more information, please view on a desktop device.
Logo for Pre-trained Deformable DETR ImageNet weights


Pre-trained deformable_detr weights trained on ImageNet to facilitate transfer learning using TAO Toolkit.



Latest Version



July 21, 2023


350.51 MB

TAO Pretrained Commercial Backbone for Deformable DETR

What is Train Adapt Optimize (TAO) Toolkit?

Train Adapt Optimize (TAO) Toolkit is a Python-based AI toolkit for taking purpose-built pre-trained AI models and customizing them with your own data. TAO adapts popular network architectures and backbones to your data, allowing you to train, fine tune, prune, and export highly optimized and accurate AI models for edge deployment.

Pre-trained models accelerate the AI training process and reduce costs associated with large scale data collection, labeling, and training models from scratch. Transfer learning with pre-trained models can be used for AI applications in smart cities, retail, healthcare, industrial inspection, and more.

Build end-to-end services and solutions for transforming pixels and sensor data to actionable insights using TAO DeepStream SDK and TensorRT. These models are suitable for object detection, classification, and segmentation.

Deformable-DETR Based Object Detection

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 places bounding boxes around the object. This model card contains pretrained weights that may be used as a starting point with the Deformable-DETR object detection networks in Train Adapt Optimize (TAO) Toolkit to facilitate transfer learning.

It is trained on the ImageNet-1K. Following backbones are supported with Deformable-DETR networks.

Supported Backbone:

  • resnet_50
  • gc_vit_xxtiny / gc_vit_xtiny / gc_vit_tiny / gc_vit_small / gc_vit_base / gc_vit_large / gc_vit_large_384

Model Versions

  • gcvit_xxtiny_imagenet1k - ImageNet1K pre-trained GCViT-xxTiny model for finetune.
  • gcvit_xtiny_imagenet1k - ImageNet1K pre-trained GCViT-xTiny model for finetune.
  • gcvit_tiny_imagenet1k - ImageNet1K pre-trained GCViT-Tiny model for finetune.
  • gcvit_small_imagenet1k - ImageNet1K pre-trained GCViT-Small model for finetune.
  • gcvit_base_imagenet1k - ImageNet1K pre-trained GCViT-Base model for finetune.
  • gcvit_large_imagenet1k - ImageNet1K pre-trained GCViT-Large model for finetune.
  • gcvit_large_imagenet22k_384 - ImageNet22k pre-trained GCViT-Large model for finetune.

Instructions to Use Pretrained Backbone Models with TAO

To use these models as pretrained backbone weights for transfer learning, use the snippet below as a template for the model and train component of the experiment spec file to train a Deformable DETR model. For more information on the experiment spec file, please refer to the TAO Toolkit User Guide.

  pretrained_backbone_path: /path/to/the/resnet50.pth
  backbone: resnet_50
  train_backbone: True
  num_feature_levels: 4
  dec_layers: 6
  enc_layers: 6
  num_queries: 300
  with_box_refine: True
  dropout_ratio: 0.3

Other TAO Pre-trained Models


This work is licensed under the Creative Commons Attribution NonCommercial ShareAlike 4.0 License (CC-BY-NC-SA-4.0). To view a copy of this license, please visit this link, or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.

Technical blogs

Suggested reading

Ethical AI

NVIDIA’s platforms and application frameworks enable developers to build a wide array of AI applications. Consider potential algorithmic bias when choosing or creating the models being deployed. Work with the model’s developer to ensure that it meets the requirements for the relevant industry and use case; that the necessary instruction and documentation are provided to understand error rates, confidence intervals, and results; and that the model is being used under the conditions and in the manner intended.