Viet-Anh on Software Logo

What is: Dynamic SmoothL1 Loss?

SourceDynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

Dynamic SmoothL1 Loss (DSL) is a loss function in object detection where we change the shape of loss function to gradually focus on high quality samples:

DSL(x,β_now)=0.5x2/β_now, if x<β_now,\text{DSL}\left(x, \beta\_{now}\right) = 0.5|{x}|^{2}/\beta\_{now}, \text{ if } |x| < \beta\_{now}\text{,} DSL(x,β_now)=x0.5β_now, otherwise\text{DSL}\left(x, \beta\_{now}\right) = |{x}| - 0.5\beta\_{now}\text{, otherwise}

DSL will change the value of β_now\beta\_{now} according to the statistics of regression errors which can reflect the localization accuracy. It was introduced as part of the Dynamic R-CNN model.