STIR
Scaled and Translated Image Recognition
ImagesCC BY 4.0Introduced 2022-11-15
While convolutions are known to be invariant to (discrete) translations, scaling continues to be a challenge and most image recognition networks are not invariant to them. To explore these effects, we have created the Scaled and Translated Image Recognition (STIR) dataset. This dataset contains objects of size s \in \[17, 64\], each randomly placed in a pixel image.
Related Benchmarks
Stirling-HQ (FG2018 3D face reconstruction challenge)/3D/Mean Reconstruction Error (mm)Stirling-HQ (FG2018 3D face reconstruction challenge)/3D Face Modelling/Mean Reconstruction Error (mm)Stirling-HQ (FG2018 3D face reconstruction challenge)/3D Face Reconstruction/Mean Reconstruction Error (mm)Stirling-HQ (FG2018 3D face reconstruction challenge)/Face Reconstruction/Mean Reconstruction Error (mm)Stirling-HQ (FG2018 3D face reconstruction challenge)/Facial Recognition and Modelling/Mean Reconstruction Error (mm)Stirling-LQ (FG2018 3D face reconstruction challenge)/3D/Mean Reconstruction Error (mm)Stirling-LQ (FG2018 3D face reconstruction challenge)/3D Face Modelling/Mean Reconstruction Error (mm)Stirling-LQ (FG2018 3D face reconstruction challenge)/3D Face Reconstruction/Mean Reconstruction Error (mm)Stirling-LQ (FG2018 3D face reconstruction challenge)/Face Reconstruction/Mean Reconstruction Error (mm)Stirling-LQ (FG2018 3D face reconstruction challenge)/Facial Recognition and Modelling/Mean Reconstruction Error (mm)