AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Uncertainty Inspired RGB-D Saliency Detection

Jing Zhang, Deng-Ping Fan, Yuchao Dai, Saeed Anwar, Fatemeh Sadat Saleh, Sadegh Aliakbarian, Nick M. Barnes

TPAMI 2021 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Disentangled and discrete latent models · 146 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 early fusion of latent z instead of late fusion (ablation)Ours CVAE · LFSDS-measure: .868 -> .839Improved
add early fusion of latent z instead of middle fusion (ablation)Ours CVAE · NJU2KS-measure: .902 -> .897Improved
add saliency consensus module instead of averaging predictions (ablation)Ours CVAE · NJU2KS-measure: .902 -> .900Improved
add saliency consensus module instead of averaging z (ablation)Ours CVAE · NJU2KS-measure: .902 -> .901Improved
add hybrid CVAE+GSNN loss instead of CVAE loss only (ablation)Ours CVAE · NJU2KS-measure: .902 -> .900Improved
add smoothness loss (ablation)Ours CVAE · SSBS-measure: .898 -> .885Improved
add structure-aware loss instead of cross-entropy (ablation)Ours CVAE · LFSDS-measure: .868 -> .851Improved
add structure-aware loss instead of cross-entropy (ablation)Ours CVAE · SIPF-measure: .877 -> .856Improved
replace raw depth with HHA encoding (ablation)Ours CVAE · SIPS-measure: .883 -> .870No clear effect
add KL annealing (ablation)Ours CVAE · NJU2KS-measure: .902 -> .900No clear effect
add KL annealing (ablation)Ours CVAE · RGB-D test setsprediction varianceImproved
increase latent dimension from 3 up to 32 (ablation)Ours CVAE · six RGB-D datasetsMAEImproved

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