E-ELAN extends the concept of ELAN 2.1.1 Why was ELAN born? The purpose of ELAN is to solve the problem that the convergence of the deep model will gradually d… See more

Model scaling refers to width (channel) or depth (number of layers) scaling. The authors found that when depth scaling was done in concatenation based models (computati… See more

BoF refers to the techniques that improves the model performance without increasing the training cost. YOLOv7 employed a few different BoFs 1. Re-parameterization Convolutio… See more

3.1 Cross Stage Partial Network
Cross Stage Partial Network is a new way to architect convolutional neural networks that reduces computation for various CNN networks … See more