AI Paper Atlas

Research map / Image generation / Super-resolution

Benchmarking Burst Super-Resolution for Polarization Images: Noise Dataset and Analysis

Inseung Hwang, Kiseok Choi, Hyunho Ha, Min H. Kim

ICCV 2025 · Image generation › Super-resolution · approach: Efficient lightweight super-resolution · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train MFIR on polarization data instead of using RGB-trained modelMFIR · Real polar dataset DoLPPSNR: 23.76 -> 22.00Hurt
train BSRT on polarization data instead of using RGB-trained modelBSRT · Real polar dataset DoLPPSNR: 23.03 -> 25.17Improved
train Burstormer on polarization data instead of using RGB-trained modelBurstormer · Real polar dataset DoLPPSNR: 24.90 -> 25.84Improved
train FBAnet on polarization data instead of using RGB-trained modelFBAnet · Real polar dataset DoLPPSNR: 24.28 -> 26.92Improved
train BurstM on polarization data instead of using RGB-trained modelBurstM · Real polar dataset DoLPPSNR: 22.78 -> 23.67Improved
train MFIR on polarization data instead of using RGB-trained modelMFIR · Real polar dataset AoLPPSNR: 16.67 -> 15.74Hurt
train BSRT on polarization data instead of using RGB-trained modelBSRT · Real polar dataset AoLPPSNR: 16.14 -> 17.06Improved
train Burstormer on polarization data instead of using RGB-trained modelBurstormer · Real polar dataset AoLPPSNR: 16.95 -> 17.36Improved
train FBAnet on polarization data instead of using RGB-trained modelFBAnet · Real polar dataset AoLPPSNR: 16.14 -> 18.14Improved
train BurstM on polarization data instead of using RGB-trained modelBurstM · Real polar dataset AoLPPSNR: 16.64 -> 16.32Hurt
train MFIR on polarization data instead of using RGB-trained modelMFIR · Synthetic polar dataset DoLPPSNR: 16.79 -> 21.26Improved
train BSRT on polarization data instead of using RGB-trained modelBSRT · Synthetic polar dataset DoLPPSNR: 20.70 -> 20.31Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy