SOTAVerified

Defocus Estimation

Papers

Showing 1–8 of 8 papers

TitleStatusHype
Multi-task Learning for Monocular Depth and Defocus Estimations with Real ImagesCode0
Single image deep defocus estimation and its applicationsCode0
Deep Defocus Map Estimation Using Domain AdaptationCode0
DeFusionNET: Defocus Blur Detection via Recurrently Fusing and Refining Multi-Scale Deep Features—0
Enhancing Diversity of Defocus Blur Detectors via Cross-Ensemble Network—0
Defocus Blur Detection via Multi-Stream Bottom-Top-Bottom Fully Convolutional Network—0
A Unified Approach of Multi-scale Deep and Hand-crafted Features for Defocus EstimationCode0
Convergence Analysis of MAP based Blur Kernel Estimation—0
Show:102550

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DMENet (BDCS)Blur Segmentation Accuracy87.35—Unverified
2DHDEBlur Segmentation Accuracy83.73—Unverified
3DeFusionNetMAE0.12—Unverified
4BTBNet (3S + FNet + RRNet)MAE0.11—Unverified
5CENetMAE0.06—Unverified