A question between yolov5s-p5 and the parameters imgsz

[jameschengds]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

as we know the first input feature map in yolov5s-p5 model is 640, and when I set imgsz to 1280, the training is better than 640. But the input of the first feature map is constant. My question is, how does the model input the 1280*1280 image to the network when I train with the p5 model with imgsz=1280? thank you!

Additional

No response

[github-actions]

👋 Hello @jameschengds, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset image examples and training logs, and verify you are following our Tips for Best Training Results.

Requirements

Python>=3.8.0 with all requirements.txt installed including PyTorch>=1.8. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training, validation, inference, export and benchmarks on macOS, Windows, and Ubuntu every 24 hours and on every commit.

Introducing YOLOv8 🚀

We're excited to announce the launch of our latest state-of-the-art (SOTA) object detection model for 2023 - YOLOv8 🚀!

Designed to be fast, accurate, and easy to use, YOLOv8 is an ideal choice for a wide range of object detection, image segmentation and image classification tasks. With YOLOv8, you'll be able to quickly and accurately detect objects in real-time, streamline your workflows, and achieve new levels of accuracy in your projects.

Check out our YOLOv8 Docs for details and get started with:

pip install ultralytics

[glenn-jocher]

Hello! Great question and thanks for doing a thorough search before asking! 😊

When you set imgsz to 1280 for training while using a YOLOv5s-p5 model, the image initially gets resized to 1280x1280 before being processed by the network. Despite the design of the first convolution layer in the network being optimized for 640x640 inputs, this resizing allows the network to process higher resolution images, effectively capturing more details and potentially improving the detection performance.

So, essentially, the model adjusts to handle the larger image input by resizing it as per the specified imgsz. This is why you're observing better training results at a higher resolution. The added detail from the larger input size can sometimes help in capturing more refined features which are beneficial during training.

Happy detecting! 🚀