AI Paper Atlas

Research map / Detection and segmentation / CNN features

Rethinking the Defocus Blur Detection Problem and A Real-Time Deep DBD Model

Ning Zhang, Junchi Yan

ECCV 2020 · Detection and segmentation › CNN features · approach: RGB-D salient object detection · 15 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 hard mining to BCE+self-supervision (ablation)BCE+self-supervision U-Net · CUHKF-measure: 0.933 -> 0.944Improved
add hard mining to BCE+self-supervision (ablation)BCE+self-supervision U-Net · CUHKMAE: 0.060 -> 0.053Improved
add hard mining to BCE+self-supervision (ablation)BCE+self-supervision U-Net · DUTF-measure: 0.811 -> 0.828Improved
add hard mining to BCE+self-supervision (ablation)BCE+self-supervision U-Net · DUTMAE: 0.127 -> 0.115Improved
add self-supervision loss to BCE (ablation)BCE U-Net · CUHKF-measure: 0.927 -> 0.933Improved
add self-supervision loss to BCE (ablation)BCE U-Net · CUHKMAE: 0.066 -> 0.060Improved
add self-supervision loss to BCE (ablation)BCE U-Net · DUTF-measure: 0.793 -> 0.811Improved
add self-supervision loss to BCE (ablation)BCE U-Net · DUTMAE: 0.140 -> 0.127Improved
add image gradient input to U-Net (ablation)BCE U-Net · CUHKF-measure: 0.918 -> 0.927Improved
add image gradient input to U-Net (ablation)BCE U-Net · DUTF-measure: 0.754 -> 0.793Improved
add image gradient input to U-Net (ablation)BCE U-Net · FPSspeed: 112.35 -> 58.82Hurt
vary self-supervision weight lambda_s from 0.0 to 0.5 (ablation)BCE+self-supervision · CUHK / DUTF-measure and MAE: CUHK F 0.927 -> 0.933 at 0.2; DUT F 0.793 -> 0.811 at 0.2No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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