Logistics Object Detection Dataset and Pre-Trained Model by Large Benchmark Datasets

Images

99238 images

Explore Dataset

Logistics Pre-trained Object Detection Model

Pre-trained models are trained on large datasets until they achieve good generalization, meaning they can recognize patterns effectively. "pre-trained" indicates that the model has already undergone training on a substantial dataset, often a generic one, and is ready for fine-tuning on a specific task with a smaller dataset. The Logistics Object Detection Base Model is a pre-trained model hosted on Roboflow Universe, created to be a strong starting point for custom training on logistics-specific object detection tasks. This model is built on a dataset of 99,238 images across 20 logistics-focused classes, collected from various projects on Roboflow Universe. Part of this dataset was auto-labeled using the Autodistill DETIC tool from Roboflow, helping to achieve a mean Average Precision (mAP) of 76%.

Classes:

Barcode, QR Code
Car, Truck, Van
Cardboard Box, Wood Pallet, Freight Container
Fire, Smoke
Forklift
Gloves, Helmet, Safety Vest
Ladder
License Plate
Person
Road Sign, Traffic Cone, Traffic Light

Current Status: The model has achieved a mAP of 76%, marking its readiness as a checkpoint for further custom training. It aims to shorten the development cycle, facilitating better model performance in specific logistics scenarios.

Trained Model API

This project has a trained model available that you can try in your browser and use to get predictions via our Hosted Inference API and other deployment methods.

YOLOv8

This project has a YOLOv8 model checkpoint available for inference with Roboflow Deploy. YOLOv8 is a new state-of-the-art real-time object detection model.

Cite This Project

If you use this dataset in a research paper, please cite it using the following BibTeX:

@misc{
                            logistics-sz9jr_dataset,
                            title = { Logistics Dataset },
                            type = { Open Source Dataset },
                            author = { Large Benchmark Datasets },
                            howpublished = { \url{ https://universe.roboflow.com/large-benchmark-datasets/logistics-sz9jr } },
                            url = { https://universe.roboflow.com/large-benchmark-datasets/logistics-sz9jr },
                            journal = { Roboflow Universe },
                            publisher = { Roboflow },
                            year = { 2023 },
                            month = { oct },
                            note = { visited on 2024-04-27 },
                            }