AI Paper Atlas

Research map / Detection and segmentation / CNN features

MobileSal: Extremely Efficient RGB-D Salient Object Detection

Yu-Huan Wu, Yun Liu, Jun Xu, Jia-Wang Bian, Yu-Chao Gu, Ming‐Ming Cheng

TPAMI 2021 · Detection and segmentation › CNN features · approach: RGB-D salient object detection · 184 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 implicit depth restoration (IDR) branch, fusing at coarsest level (ablation)MobileSal · NJU2K testF-measure max: 0.896 -> 0.906Improved
fuse RGB and depth at the coarsest level only instead of all levels (ablation)MobileSal with IDR · NJU2K testF-measure max: 0.899 -> 0.906Improved
fuse only at the coarsest level instead of all levels (speed) (ablation)MobileSal · speedFPS: 260 -> 450 FPSImproved
use two-stream fusion instead of early fusion single stream (ablation)MobileSal with IDR · NJU2K testF-measure max: 0.900 -> 0.906Improved
add CPR module to RGB backbone (ablation)MobileSal · NJU2K testF-measure max: 0.852 -> 0.887Improved
add depth stream (multiplication fusion) (ablation)RGB+CPR · NJU2K testF-measure max: 0.887 -> 0.894Improved
add CMF module (ablation)RGB+CPR+Depth · NJU2K testF-measure max: 0.894 -> 0.906Improved
element-wise multiplication instead of addition or concatenation for initial fusion (ablation)CMF · NJU2K testF-measure max: 0.897 -> 0.906Improved
keep RGB-D feature T in CMF fusion (ablation)CMF · NJU2K testF-measure max: 0.895 -> 0.906Improved
keep RGB attention vector v in CMF fusion (ablation)CMF · NJU2K testF-measure max: 0.902 -> 0.906Improved
use compact dilation rates 1, 2, 3 instead of sparse 1, 3, 6 (ablation)CPR · NJU2K testF-measure max: 0.903 -> 0.906Improved
use SSIM loss instead of L1/L2 loss for IDR (ablation)IDR · NJU2K testF-measure max: 0.5%/0.3% lower with L1/L2Improved

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