AI Paper Atlas

Research map / Detection and segmentation / CNN features

Divide-and-Conquer: Confluent Triple-Flow Network for RGB-T Salient Object Detection

Hao Tang, Zechao Li, Dong Zhang, Shengfeng He, Jinhui Tang

TPAMI 2024 · Detection and segmentation › CNN features · approach: RGB-D salient object detection · 289 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 the full modality-induced feature modulator (MFM) (ablation)single-flow baseline · VT5000S-measure: 0.876 -> 0.884Improved
add CFE to MFM that has only AFF (ablation)MFM · VT5000S-measure: 0.881 -> 0.884Improved
add AFF to MFM that has only CFE (ablation)MFM · VT5000S-measure: 0.883 -> 0.884Improved
add a second flow (flow x2) (ablation)baseline with MFM · VT5000S-measure: 0.884 -> 0.890Improved
add RASPM (ablation)triple-flow decoders · VT821S-measure: 0.883 -> 0.885Improved
add MDAM (ablation)flow x2 + RASPM · VT5000S-measure: 0.889 -> 0.894Improved
add DW to MDAM (ablation)MDAM · VT5000S-measure: 0.892 -> 0.894Improved
add DoE to MDAM (ablation)MDAM · VT5000S-measure: 0.893 -> 0.894Improved
use RASPM instead of ASPP (ablation)decoder module · VT5000S-measure: 0.889 -> 0.894Improved
add atrous convolution to RASPM (ablation)RASPM · VT5000S-measure: 0.893 -> 0.894Improved
add weighted IoU loss to weighted BCE loss (ablation)ConTriNet · VT5000F-measure: 0.840 -> 0.863Improved
add weighted BCE loss to weighted IoU loss (ablation)ConTriNet · VT5000F-measure: 0.857 -> 0.863Improved

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