AIbase
EN
Home
Detr Resnet 101
D

Detr Resnet 101

Developed by Xenova
End-to-end object detection model based on Transformer architecture with ResNet-101 feature extractor
Object Detection
Transformers
#Object Detection#ONNX Format#Web Adaptation
Downloads 216
Release Time : 5/3/2023

Model Overview

DETR is an end-to-end object detection model based on Transformer, eliminating the need for traditional hand-designed components (such as non-maximum suppression). This version uses ResNet-101 as the backbone network for feature extraction.

Model Features

End-to-end Detection
Eliminates the need for hand-designed components in traditional object detection pipelines (such as anchor generation or non-maximum suppression)
Transformer Architecture
Utilizes attention mechanisms to directly model global relationships in object detection tasks
ResNet-101 Backbone
Uses the well-established ResNet-101 network for feature extraction, ensuring feature quality
ONNX Format Support
Converted to ONNX weight format for easy deployment in web environments

Model Capabilities

Image Object Detection
Multi-category Object Recognition
Bounding Box Prediction

Use Cases

Computer Vision
Smart Surveillance
Real-time detection of multiple target objects in surveillance footage
Can identify various targets such as people and vehicles
Autonomous Driving
Object detection and localization in road scenes
Can detect pedestrians, traffic signs, other vehicles, etc.
E-commerce
Product Recognition
Automatically identify items in product images
Can be used for inventory management or image search
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
EnglishįŽ€äŊ“中文įšéĢ”中文ãĢãģんご
Š 2025AIbase