Scaled-YOLOv4: Scaling Cross Stage Partial Network

by   Chien-Yao Wang, et al.

We show that the YOLOv4 object detection neural network based on the CSP approach, scales both up and down and is applicable to small and large networks while maintaining optimal speed and accuracy. We propose a network scaling approach that modifies not only the depth, width, resolution, but also structure of the network. YOLOv4-large model achieves state-of-the-art results: 55.4 V100, while with the test time augmentation, YOLOv4-large achieves 55.8 (73.2 AP50). To the best of our knowledge, this is currently the highest accuracy on the COCO dataset among any published work. The YOLOv4-tiny model achieves 22.0 using TensorRT, batch size = 4 and FP16-precision the YOLOv4-tiny achieves 1774 FPS.


EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Convolutional Neural Networks (ConvNets) are commonly developed at a fix...

YOLOv4: Optimal Speed and Accuracy of Object Detection

There are a huge number of features which are said to improve Convolutio...

EfficientDet: Scalable and Efficient Object Detection

Model efficiency has become increasingly important in computer vision. I...

Optimizing Anchor-based Detectors for Autonomous Driving Scenes

This paper summarizes model improvements and inference-time optimization...

SplitNet: Divide and Co-training

The width of a neural network matters since increasing the width will ne...

Smart Inference for Multidigit Convolutional Neural Network based Barcode Decoding

Barcodes are ubiquitous and have been used in most of critical daily act...

Code Repositories


Scaled-YOLOv4: Scaling Cross Stage Partial Network

view repo

Please sign up or login with your details

Forgot password? Click here to reset