search

ultralytics / yolov3

YOLOv3 in PyTorch > ONNX > CoreML > TFLite
https://docs.ultralytics.com
GNU Affero General Public License v3.0
10.22k stars 3.45k forks source link

Why the yolov4's inference speed is much slower than yolov3-spp? #1296

Closed buaazsj closed 4 years ago

buaazsj commented 4 years ago

hi, Why the yolov4's inference speed is much slower than yolo-spp? single P40 GPU

  1. yolov4: 65.3ms per 512*512 image
  2. yolov3-spp : about 20ms per 512*512 image

image image

github-actions[bot] commented 4 years ago

Hello @buaazsj, thank you for your interest in our work! Ultralytics has publicly released YOLOv5 at https://github.com/ultralytics/yolov5, featuring faster, lighter and more accurate object detection. YOLOv5 is recommended for all new projects.

To continue with this repo, please visit our Custom Training Tutorial to get started, and see our Google Colab Notebook, Docker Image, and GCP Quickstart Guide for example environments.

If this is a bug report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom model or data training question, please note that Ultralytics does not provide free personal support. As a leader in vision ML and AI, we do offer professional consulting, from simple expert advice up to delivery of fully customized, end-to-end production solutions for our clients, such as:

For more information please visit https://www.ultralytics.com.

github-actions[bot] commented 4 years ago

This issue is stale because it has been open 30 days with no activity. Remove Stale label or comment or this will be closed in 5 days.

laizaparizotto commented 4 years ago

I also have the same question. My yolov4 is with an inference time of 100 milliseconds per image, while the v3 is about 17 milliseconds.

glenn-jocher commented 11 months ago

@laizaparizotto the performance difference between YOLOv4 and YOLOv3-SPP is likely due to YOLOv4's increased model complexity and larger number of layers. YOLOv4 was designed to optimize accuracy, sacrificing some speed. However, you can improve YOLOv4's performance by using model quantization or running the model on a faster GPU. Keep in mind that these inference times are not standardized and may vary across different hardware configurations.