AI Paper Atlas

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

Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation

Alexander Kolesnikov, Christoph H. Lampert

ECCV 2016 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised semantic segmentation · 786 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
use GWRP instead of global max pooling for present classes (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 47.3 -> 50.7Improved
use GWRP instead of global average pooling for present classes (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 45.1 -> 50.7Improved
add expansion and constrain losses to seeding loss (full SEC vs L_seed only) (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 49.2 -> 50.7Improved
add constrain-to-boundary loss to seeding loss (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 49.4 -> 50.7Improved
add expansion loss to seeding loss alone (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 49.2 -> 45.7Hurt
add constrain loss to seeding loss alone (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 49.2 -> 49.4No clear effect
add seeding loss to expansion and constrain losses (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 17.2 -> 50.7Improved
add seeding loss to expansion loss alone (ablation)SEC · PASCAL VOC 2012 valmIoU (%): 27.8 -> 45.7Improved
use large field-of-view network (378x378) instead of small (211x211) (ablation)SEC · PASCAL VOC 2012 valmIoU (%)Improved
train SEC with image-level labels versus prior best weakly supervised (val)SEC · PASCAL VOC 2012 valmIoU (%): 36.6 -> 50.7Improved
train SEC with image-level labels versus prior best weakly supervised (test)SEC · PASCAL VOC 2012 testmIoU (%): 38.0 -> 51.7Improved

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