DETR-ResNet50: Object Detection

DETR-ResNet50 is an efficient object detection model that combines DETR (DEtection TRansformers) with the ResNet-50 backbone network. DETR is an end-to-end object detection framework based on Transformer architecture, utilizing self-attention mechanisms to detect objects in images without relying on traditional region proposals. ResNet-50 serves as the backbone for feature extraction, leveraging residual connections to effectively learn multi-level features of the image, enhancing detection capability. DERT-ResNet50 strikes a balance between high detection accuracy and reduced complexity, making it suitable for object detection tasks in complex scenes, with applications in autonomous driving, video surveillance, and real-time object detection.

Source model

  • Input shape: 480x480
  • Number of parameters: 39.60M
  • Model size: 158.03M
  • Output shape: 1x100x92, 1x100x4

Source model repository: DETR-ResNet50

Performance Reference

Please search model by model name in Model Farm

Inference & Model Conversion

Please search model by model name in Model Farm

License

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support