AI Paper Atlas

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

Soft Proposal Networks for Weakly Supervised Object Localization

Yi Zhu, Yanzhao Zhou, Qixiang Ye, Qiang Qiu, Jianbin Jiao

ICCV 2017 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised object localization · 155 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
upgrade VGG16 to SPN by adding Soft Proposal moduleVGG16 · VOC2007 test pointing (All)pointing accuracy: 80.0 -> 87.5Improved
upgrade VGG16 to SPN by adding Soft Proposal moduleVGG16 · VOC2007 test pointing (Diff.)pointing accuracy: 66.8 -> 78.1Improved
upgrade GoogLeNet to SPN by adding Soft Proposal moduleGoogLeNet · VOC2007 test pointing (All/Diff.)pointing accuracy: 85.1/72.3 -> 88.2/79.1Improved
upgrade CNN-S to SPNCNN-S · VOC2007 test pointing (All/Diff.)pointing accuracy: 78.7/61.7 -> 81.8/66.7Improved
use weakly supervised soft proposal instead of supervised RPNSPN · VOC2007 test proposal qualityObject Energy: 63.3 -> 62.2No clear effect
use soft proposal instead of RPN for proposal timeSPN · VOC2007 test proposal qualitytime per image (ms): 10.5 -> 0.9Improved
use soft proposal instead of Selective SearchSPN · VOC2007 test proposal qualityObject Energy: 53.7 -> 62.2Improved
upgrade pre-trained VGG16 to SPN, pointing with predictionVGG16 · VOC2012 val pointing with predictionmAP: 77.1 -> 82.9Improved
upgrade pre-trained VGG16 to SPN, pointing with predictionVGG16 · COCO2014 val pointing with predictionmAP: 49.2 -> 55.3Improved
use SP-VGGNet for bounding box localizationVGG16 · VOC2007 trainval positive images CorLocCorLoc mean: 55.1 -> 60.6Improved
add Soft Proposal module to GoogLeNetGAPGoogLeNetGAP · ILSVRC2014 val classificationtop-1/top-5 error: 35.0/13.2 -> 33.5/12.7Improved
add Soft Proposal module to GoogLeNetGAP then fine-tuneGoogLeNetGAP · VOC2007 test classification (fine-tuned)mAP: 83.4 -> 84.2Improved

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