AI Paper Atlas

Research map / Detection and segmentation / CNN features

Better to Follow, Follow to Be Better: Towards Precise Supervision of Feature Super-Resolution for Small Object Detection

Junhyug Noh, Wonho Bae, Wonhee Lee, Jinhwan Seo, Gunhee Kim

ICCV 2019 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 276 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 content loss with SR target extractor supervision (ablation)Faster R-CNN ResNet-50 + SR · Tsinghua-Tencent 100Koverall F1: 87.5 -> 89.7Improved
use SR target extractor (matched receptive field) instead of naive targets (ablation)Faster R-CNN ResNet-50 + SR · Tsinghua-Tencent 100Koverall F1: 85.3 -> 89.7Improved
add naive supervision from base extractor (ablation)Faster R-CNN ResNet-50 + SR · Tsinghua-Tencent 100Koverall F1: 87.5 -> 85.3Hurt
add feature-level SR for small proposalsFaster R-CNN ResNet-50 · Tsinghua-Tencent 100K small objectsF1: 74.9 -> 82.2Improved
add feature-level SR for small proposalsFaster R-CNN ResNet-101 · Tsinghua-Tencent 100K small objectsF1: 75.2 -> 84.3Improved
add feature-level SR for small proposalsFaster R-CNN MobileNet · Tsinghua-Tencent 100K small objectsF1: 63.4 -> 71.0Improved
add feature-level SR for small proposalsFaster R-CNN ResNet-101 · Tsinghua-Tencent 100K large objectsF1: 92.2 -> 93.2Improved
add feature-level SR for small proposalsFaster R-CNN MobileNet · Tsinghua-Tencent 100K large objectsF1: 87.1 -> 86.5Hurt
add feature-level SR for small proposalsFaster R-CNN ResNet-101 · COCO 2017 test-devAP-S: 11.3 -> 16.2Improved
add feature-level SR for small proposalsFaster R-CNN ResNet-50 · PASCAL VOC 2007 testAP-S: 6.8 -> 10.5Improved
add feature-level SR for small proposalsFaster R-CNN MobileNet · COCO 2017 test-devAP-L: 29.2 -> 29.0No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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