AI Paper Atlas

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

Weakly-Supervised Image Annotation and Segmentation with Objects and Attributes

Zhiyuan Shi, Yongxin Yang, Timothy M. Hospedales, Tao Xiang

TPAMI 2016 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised detection · 49 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 8 material attribute annotations to weakly supervised segmentation (ablation)WS-MRF-SIBP · aPascal segmentationavg IOU (%): 23.6 -> 27.3Improved
use 64 attributes instead of 8 (ablation)WS-MRF-SIBP · aPascal segmentationavg IOU (%): 27.3 -> 28.9Improved
use 74 attributes (add 10 color attributes) instead of 64 (ablation)WS-MRF-SIBP · aPascal segmentationavg IOU (%): 28.9 -> 29.4Improved
replace WSDC with WS-MRF-SIBP without attributesweakly supervised segmentation · aPascal segmentationavg IOU (%): 18.2 -> 23.6Improved
update the model with test images (transductive mode) (ablation)WS-MRF-SIBP · LabelMe Outdoorper-pixel accuracy (%): 46.2 -> 52.5Improved
update the model with test images (transductive mode) (ablation)WS-MRF-SIBP · LabelMe Outdoorper-class accuracy (%): 23.8 -> 31.2Improved
use image-level labels predicted by an external CNN+SVM classifier instead of the model's own predictions (ablation)WS-MRF-SIBP transductive · LabelMe Outdoorper-pixel accuracy (%): 52.5 -> 48.1Hurt
replace w-LDA with WS-MRF-SIBP (IBP, non-competing factors)free annotation · aPascal free annotationAP@2: 30.7 -> 40.1Improved
replace weak labels with strong bounding-box supervisionannotation · ImageNet Attribute free annotationAP@2: 60.7 -> 65.9Improved
use weak supervision instead of strong supervision for attribute annotation given locationsWS-MRF-SIBP · aPascal annotation given locationsmAP: 42.1 -> 43.6No clear effect
use WS-MRF-SIBP instead of w-LDAtraining cost · aPascal training setseconds per iteration: 65 -> 80Hurt

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