AI Paper Atlas

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

Test Time Adaptation With Regularized Loss for Weakly Supervised Salient Object Detection

Olga Veksler

CVPR 2023 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised detection · 15 citations (OpenAlex)

Read the paper

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
adapt the base CNN on augmented copies of each test image using regularized loss and per-image hyperparametersweakly supervised SOD model [33] · DUTS, ECSSD, DUTO, PascalS, HKUIS, THURFβ: [33] -> ours+h: .753 -> .795 (DUTS); .868 -> .911 (ECSSD); .697 -> .752 (DUTO); .796 -> .807 (PascalS); .866 -> .899 (HKImproved
set λm adaptively per test image instead of using a fixed value (ablation)test-time adaptation on DUTO · DUTO (500-image sample)Fβ: fixed settings .741, .738, .752, .755, .742, .732, .728, .718, .703; adaptive .765Improved
remove border loss (ablation)test-time adaptation · DUTS, ECSSD, DUTO, PascalS, HKUIS, THURFβ: DUTS .795 -> .789; ECSSD .911 -> .910; DUTO .752 -> .748; PascalS .807 -> .800; HKUIS .899 -> .892; THUR .760 -> .755Hurt
vary sparse-CRF σ from 0.15 to 0.1 or 0.20 (ablation)test-time adaptation · DUTS, ECSSD, DUTO, PascalS, HKUIS, THURFβ: DUTS .795 -> .789/.797; ECSSD .911 -> .894/.915; DUTO .752 -> .751/.743; PascalS .807 -> .803/.803; HKUIS .899 -> .902/.Mixed
increase test-time dataset size from 16 to 32 images (ablation)test-time adaptation · ECSSDFβ: .910 -> .915Improved
increase test-time dataset size from 16 to 64 images (ablation)test-time adaptation · ECSSDFβ: .910 -> .913No clear effect
reduce adaptation to 2 or 5 epochs (ablation)test-time adaptation · ECSSDFβ: 10 epochs .910; 2 epochs .873; 5 epochs .883Hurt
increase adaptation to 20 or 30 epochs (ablation)test-time adaptation · ECSSDFβ: 10 epochs .910; 20 epochs .915; 30 epochs .915Improved
increase adaptation to 50 epochs (ablation)test-time adaptation · ECSSDFβ: 10 epochs .910 -> 50 epochs .912No clear effect
use Gaussian noise augmentation instead of small shifts (ablation)test-time adaptation · ECSSDFβ: .911 -> .854Hurt
use color perturbation instead of small shifts (ablation)test-time adaptation · ECSSDFβ: .911 -> .897Hurt
add color perturbation to small-shift augmentation (ablation)test-time adaptation · ECSSDFβ: .911 -> .909No clear effect

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