Detr-resnet-50 Open-source Object Detection Model - Achieving End-to-End Detection without Anchor Box Design

Detr Resnet 50

Developed by Xenova

An end-to-end object detection model based on Transformer architecture, eliminating the need for anchor box designs in traditional object detection

Object Detection

Transformers

#End-to-End Object Detection #Transformers.js Based #ONNX Optimized

Downloads 5,261

Release Time : 5/2/2023

Model Overview

DETR (Detection Transformer) is an object detection model based on the Transformer architecture, using ResNet-50 as its backbone network. It treats object detection as a set prediction problem, achieving an end-to-end detection process without the need for anchor box designs in traditional methods.

Model Features

End-to-End Detection

Eliminates the need for anchor box designs in traditional object detection, directly outputting detection results

Transformer-Based

Utilizes Transformer architecture for visual tasks, enabling global context understanding

Efficient Inference

Optimized through ONNX format, suitable for deployment and operation in Web environments

Model Capabilities

Object Detection

Object Localization

Multi-Class Recognition

Use Cases

Smart Surveillance

Security Monitoring

Detects targets such as people and vehicles in surveillance footage

Accurately marks various objects in the footage

Retail Analytics

Shelf Analysis

Detects product placements on retail shelves

Identifies product types and positions

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Detr Resnet 50

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 Xenova/detr-resnet-50

🚀 Quick Start

✨ Features

📦 Installation

💻 Usage Examples

Basic Usage

📚 Documentation

Demo