AI Paper Atlas

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

Self-produced Guidance for Weakly-supervised Object Localization

Xiaolin Zhang, Yunchao Wei, Guoliang Kang, Yi Yang, Thomas Huang

ECCV 2018 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised object localization · 248 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 self-produced guidance (SPG-B and SPG-C) to SPG-plainSPG-plain · ILSVRC val top-1top-1 localization error (%): 53.71 -> 51.40Improved
add self-produced guidance to SPG-plainSPG-plain · ILSVRC val top-5top-5 localization error (%): 41.81 -> 40.00Improved
add self-produced guidance to SPG-plainSPG-plain · CUB-200-2011 top-1top-1 localization error (%): 56.33 -> 53.36Improved
add self-produced guidance to SPG-plainSPG-plain · ILSVRC val (GT-known)GT-known localization error (%): 37.32 -> 35.31Improved
add two-stage cascade supervision of B1 by B2 (ablation)SPG · ILSVRC val (GT-known)GT-known localization error (%): 35.58 -> 35.31Improved
share second and third layers between B1 and B2 (ablation)SPG · ILSVRC val (GT-known)GT-known localization error (%): 36.31 -> 35.31Improved
add SPG-C auxiliary supervision (ablation)SPG · ILSVRC val (GT-known)GT-known localization error (%): 36.06 -> 35.31Improved
SPG-B alone without SPG-C versus plain network (ablation)SPG without SPG-C · ILSVRC val (GT-known)GT-known localization error (%): 37.32 -> 36.06Improved
repeat boxes from top predicted classes for top-5 (SPG*)SPG · ILSVRC val top-5top-5 localization error (%): 40.00 -> 35.05Improved
repeat boxes from top predicted classes for top-5 (SPG*)SPG · CUB-200-2011 top-5top-5 localization error (%): 42.28 -> 40.62Improved
combine with stronger classifier ResNet-50 instead of own classificationGoogLeNet-SPG · ILSVRC val top-1top-1 localization error (%): 51.40 -> 48.79Improved
combine with DPN-ensemble classifier instead of ResNet-50GoogLeNet-SPG · ILSVRC val top-1top-1 localization error (%): 48.79 -> 43.83Improved

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