The Best 320 Object Detection Tools in 2025

A table detection model based on the DETR architecture, specifically designed for extracting tables from unstructured documents

Grounding DINO is an open-set object detection model that achieves zero-shot object detection capabilities by combining the DINO detector with a text encoder.

Grounding DINO is an open-set object detection model that combines the DINO detector with grounding pre-training, enabling zero-shot object detection.

DETR is an end-to-end object detection model based on Transformer architecture, using ResNet-50 as the backbone network and trained on the COCO dataset.

DETR is an end-to-end object detection model using Transformer architecture, employing ResNet-101 as the backbone network and trained on the COCO dataset.

Document table detection model based on DETR architecture for detecting bordered and borderless tables in documents

A vision Transformer (ViT)-based object detection model trained with DETR loss function, achieving excellent performance on the COCO dataset.

YOLOS model fine-tuned on the COCO 2017 object detection dataset, utilizing Vision Transformer architecture for efficient object detection.

RT-DETR is the first real-time end-to-end object detector, achieving 53.1% AP and 108 FPS on the COCO dataset through an efficient hybrid encoder and uncertainty-minimized query selection mechanism.

The first real-time end-to-end object detector based on Transformer architecture, eliminating the need for non-maximum suppression, surpassing YOLO series in speed and accuracy

An object detection model based on YOLOv8m, specifically designed for table extraction tasks, capable of detecting both bordered and borderless tables.

Lightweight license plate detection model based on YOLOv5n, optimized for license plate recognition tasks

A table detection model based on YOLOv8s, capable of accurately identifying bordered and borderless tables in images.

RT-DETRv2 is an optimized real-time object detection model based on the RT-DETR architecture. It improves detection accuracy while maintaining real-time performance through selective multi-scale feature extraction and enhanced training strategies.

This is a fine-tuned object detection model for the fashion domain, based on the YOLOS architecture, capable of recognizing and localizing fashion items and their details.

YOLOE is an efficient, unified, and open model for object detection and segmentation, supporting various prompting mechanisms including text, visual inputs, and prompt-free paradigms, achieving real-time universal visual perception.

OmDet-Turbo is an efficient fusion-head open-vocabulary detection model based on real-time Transformer, suitable for zero-shot object detection tasks.

RT-DETR is the first real-time end-to-end object detector, achieving a balance of speed and accuracy surpassing the YOLO series through efficient hybrid encoders and query selection mechanisms

A YOLOv8s fine-tuned model specialized for locating signatures in document images

Deformable DETR is an end-to-end object detection model that improves detection performance using Transformer architecture and deformable attention mechanisms.

A deformable DETR model trained on the DocLayNet dataset for object detection tasks in document layout analysis, achieving a box mAP of 57.1.

RT-DETRv2 is an improved real-time object detection Transformer model that enhances performance through strategies such as selective multi-scale feature extraction and dynamic data augmentation.

A license plate object detection model based on the YOLOv5m architecture, excelling in license plate detection tasks.

A document layout detection model based on the DETR architecture, capable of identifying various layout elements in documents.

RT-DETR is the first real-time end-to-end object detection Transformer model, achieving efficient NMS-free detection through a hybrid encoder and query selection mechanism

TF-ID is a visual object detection model specifically designed for extracting tables and charts from academic papers, fine-tuned based on Florence-2

DETR is an end-to-end object detection model using Transformer, employing ResNet-101 as the backbone network and trained on the COCO dataset.

Conditional DETR is an improved object detection model that significantly enhances training convergence speed through conditional cross-attention mechanisms

A YOLOv8-based object detection model specifically designed for real-time detection of key patterns in stock market trading charts

A table detection model based on YOLOv8 for identifying table regions in images, supporting both bordered and borderless table detection.

YOLOv8 is the latest generation object detection model developed by Ultralytics, building on the success of previous YOLO versions with new features and improvements to further enhance performance and flexibility.

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

D-FINE is a model for object detection that achieves excellent positioning accuracy by redefining the bounding box regression task in the DETR model.

YOLOE is a real-time visual omni-model that combines object detection and visual understanding capabilities, suitable for various visual tasks.

An object detection model fine-tuned on the CPPE-5 dataset based on facebook/detr-resnet-50, specializing in personal protective equipment detection.

This DETR model has been trained end-to-end on the SKU110K object detection dataset with the number of queries set to 400, suitable for scenarios like product shelf detection.

DETR is an end-to-end object detection model based on Transformer architecture, using ResNet-50 as the backbone network and trained on the COCO dataset.

D-FINE is a powerful real-time object detection model that achieves excellent positioning accuracy by redefining the bounding box regression task in the DETR model.

D-FINE is a real-time object detection model that achieves exceptional localization accuracy by redefining the bounding box regression task.

LLMDet is a powerful open-vocabulary object detector supervised by large language models, a highlight paper at CVPR2025

YOLOv10 is a real-time end-to-end object detection model proposed by Tsinghua University, known for its efficiency and accuracy.

D-FINE is a real-time object detection model based on an improved DETR architecture, achieving exceptional localization accuracy by redefining the bounding box regression task.