Models

What is YOLOv4 Tiny?

The tiny and fast version of YOLOv4 - good for training and deployment on limited compute resources, and getting a feel for your dataset

About the model

Here is an overview of the

YOLOv4 Tiny

model:

Date of Release Nov 09, 2020
Model Type Object Detection
Architecture ResNet-D, YOLO
Framework Used Darknet
Annotation Format YOLO Darknet TXT
Stars on GitHub +

What is YOLOv4-Tiny

YOLOv4-tiny is the compressed version of YOLOv4 designed to train on machines that have less computing power. Its model weights are around 16 megabytes large, allowing it to train on 350 images in 1 hour when using a Tesla P100 GPU. YOLOv4-tiny has an inference speed of 3 ms on the Tesla P100, making it one of the fastest object detection models to exist.

YOLOv4-Tiny Architecture

YOLOv4-Tiny utilizes a couple of different changes from the original YOLOv4 network to help it achieve these fast speeds. First and foremost, The number of convolutional layers in the CSP backbone are compressed with a total of 29 pretrained convolutional layers. Additionally, the number of YOLO layers has been reduced to two instead of three and there are fewer anchor boxes for prediction.

YOLOv4 Results

YOLOv4-Tiny has comparatively competitive results with YOLOv4 given the size reduction. It achieves 40 mAP @.5 on the MS COCO dataset.

YOLOv4-Tiny Results

Further Reading over YOLOv4-Tiny

Training YOLOv4-tiny on Custom Data for Lightning Fast Object Detection: https://blog.roboflow.com/train-yolov4-tiny-on-custom-data-lighting-fast-detection/

Check out YOLOv8, defining a new state-of-the-art in computer vision

YOLOv8 is here, setting a new standard for performance in object detection and image segmentation tasks. Roboflow has developed a library of resources to help you get started with YOLOv8, covering guides on how to train YOLOv8, how the model stacks up against v5 and v7, and more.

Learn about YOLOv8

Check out YOLOv8, defining a new state-of-the-art in computer vision

YOLOv8 is here, setting a new standard for performance in object detection and image segmentation tasks. Roboflow has developed a library of resources to help you get started with YOLOv8, covering guides on how to train YOLOv8, how the model stacks up against v5 and v7, and more.

Learn about YOLOv8

Check out YOLOv8, defining a new state-of-the-art in computer vision

YOLOv8 is here, setting a new standard for performance in object detection and image segmentation tasks. Roboflow has developed a library of resources to help you get started with YOLOv8, covering guides on how to train YOLOv8, how the model stacks up against v5 and v7, and more.

Learn about YOLOv8

Check out YOLOv8, defining a new state-of-the-art in computer vision

YOLOv8 is here, setting a new standard for performance in object detection and image segmentation tasks. Roboflow has developed a library of resources to help you get started with YOLOv8, covering guides on how to train YOLOv8, how the model stacks up against v5 and v7, and more.

Learn about YOLOv8

Model Performance

Explore this model on Roboflow

Deploy YOLOv4 Tiny to production

Roboflow offers a range of SDKs with which you can deploy your model to production.

No items found.

YOLOv4 Tiny Annotation Format

YOLOv4 Tiny

uses the

uses the

YOLO Darknet TXT

annotation format. If your annotation is in a different format, you can use Roboflow's annotation conversion tools to get your data into the right format.

Convert data between formats

Label data automatically with YOLOv4 Tiny

You can automatically label a dataset using

YOLOv4 Tiny

with help from Autodistill, an open source package for training computer vision models. You can label a folder of images automatically with only a few lines of code. Below, see our tutorials that demonstrate how to use

YOLOv4 Tiny

to train a computer vision model.

No items found.

Deploy a computer vision model today

Join 100k developers curating high quality datasets and deploying better models with Roboflow.

Get started
MANAGING over 100 million images for companies of all sizes

Join over 250,000 developers managing computer vision data on Roboflow.

VentureBeatTechCrunchInteresting EngineeringInternational Business TimesU.S. News & World ReportYahoo Finance