AI Paper Atlas

Research map / Image generation / Super-resolution

Reference-based Super-Resolution via Image-based Retrieval-Augmented Generation Diffusion

Byeonghun Lee, Hyunmin Cho, Hong Gyu Choi, Soo Min Kang, Iljun Ahn, Kyong Hwan Jin

ICCV 2025 · Image generation › Super-resolution · approach: Diffusion-based super-resolution · 3 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
add hashing-based reference matching in training (vs random pairing) (ablation)iRAG · RealSRPSNR: PSNR from 23.957 to 23.614 without it, SSIM from 0.715 to 0.698Improved
add the reference restoration module (ablation)iRAG · RealSRPSNR: 23.957 -> 23.374 when removed, SSIM 0.715 -> 0.680Improved
add hashing-based retrieval at inference (vs random patch selection) (ablation)iRAG · RealSRPSNR: 23.957 -> 23.709 when removed, SSIM 0.715 -> 0.708Improved
raise loss-balancing weight alpha from 1.0 to 1.3 (ablation)iRAG · DIV2K-ValPSNR vs CLIP-IQA: PSNR 22.323 -> 22.007, CLIP-IQA 0.702 -> 0.716Mixed
train hash on real plus generated patches instead of real only (ablation)hash retrieval · CUFED hit rate R1hit rate (%): 45 -> 52Improved
train hash on generated patches only instead of real (ablation)hash retrieval · CUFED hit rate R1hit rate (%): 45 -> 40Hurt
augment limited real set (1K) with generated patches (ablation)hash retrieval · CUFED hit rate R1hit rate (%): 48 -> 52Improved
use 16-bit hash codes instead of VGG feature matchingreference retrieval · Retrieval costmatching time (ms): 4338.2 ms -> 67.8 msImproved
use hash-selected references instead of generated referencesCoSeR · CoSeR reference adaptationMUSIQ: 56.419 -> 57.686Improved
use hash references at training and evaluation instead of random (ablation)TTSR · CUFEDPSNR: 24.74 -> 25.65Improved

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