AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Modeling the Background for Incremental and Weakly-Supervised Semantic Segmentation

Fabio Cermelli, Massimiliano Mancini, Samuel Rota Bulò, Elisa Ricci, Barbara Caputo B

TPAMI 2021 · Detection and segmentation › Weakly supervised detection and segmentation · approach: CAM refinement and prototypes · 16 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 modified cross-entropy that sums old-class and background probability (ablation)LwF · VOC 15-1 overlappedmIoU all: 1.8 -> 9.9Improved
add modified distillation loss (KD) (ablation)LwF + CE · VOC 15-1 overlappedmIoU all: 9.9 -> 27.2Improved
add classifier initialization from background weights (ablation)LwF + CE + KD · VOC 15-1 overlappedmIoU all: 27.2 -> 29.7Improved
add classifier initialization (ablation)LwF + CE + KD · VOC 19-1 overlappedmIoU novel class: 11.9 -> 22.1Improved
add modified distillation loss (KD) (ablation)LwF + CE · VOC 19-1 overlappedmIoU all: 55.2 -> 63.3Improved
add modified KD (ablation)LwF + CE · VOC 15-5 overlappedmIoU all: 57.0 -> 66.3Improved
add classifier initialization (ablation)LwF + CE + KD · VOC 15-5 overlappedmIoU all: 66.3 -> 69.0Improved
replace standard LwF with background-aware MiBLwF · VOC 15-1 overlappedmIoU all: 1.8 -> 29.7Improved
model unlabeled pixels as background-or-annotated-classes instead of only background (ablation)PCE point supervision · VOC point supervisionmIoU: 38.8 -> 46.7Improved
add unlabeled-pixel loss to partial cross-entropyPCE · VOC scribble supervision (w/o CRF)mIoU: 69.5 -> 72.3Improved
add unlabeled-pixel loss to partial cross-entropyPCE · VOC scribble supervision (w/ CRF)mIoU: 72.8 -> 75.1Improved
add unlabeled-pixel loss to partial cross-entropyPCE · ADE20K point supervision (our protocol)mIoU: 22.4 -> 22.9Improved

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