Rediscovery of SSIM index in image reconstruction. SSIM as a loss function.

SEBO KIM
SEBO KIM
Nov 1 · 4 min read

Default loss function in encoder-decoder based image reconstruction had been L2 loss. Previously, Caffe only provides L2 loss as a built-in loss layer. Generally, L2 loss makes reconstructed image blurry because minimizing L2 loss means maximizing log-likelihood of Gaussian. As you know Gaussian is unimodal.

L1 gains a popularity over L2 because it tends to create less blurry images. However, using either L1 or L2 loss in learning takes enormous time to converge. Both losses are pointwise, error is…

Keep the story going. Sign up for an extra free read.

You've completed your member preview for this month, but when you sign up for a free Medium account, you get one more story.
Already have an account? Sign in

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade