AI Paper Atlas

Research map / Detection and segmentation / CNN features

Learning Surrogates via Deep Embedding

Yash Patel, Tomáš Hodaň, Jiří Matas

ECCV 2020 · Detection and segmentation › CNN features · approach: Efficient DETR query design · 9 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
post-tune with LS-ED instead of cross-entropy onlyTPS-ResNet-BiLSTM-Attn · IIIT-5Kaccuracy: 84.300 -> 86.300Improved
post-tune with LS-ED instead of cross-entropy onlyResNet-BiLSTM-Attn (no TPS) · ICDAR'15accuracy: 78.520 -> 78.410Hurt
post-tune with LS-EDTPS-ResNet-BiLSTM-Attn · SVTaccuracy: 87.172 -> 86.708Hurt
post-tune with LS-EDTPS-ResNet-BiLSTM-Attn · CUTEaccuracy: 73.171 -> 74.216Improved
post-tune with LS-EDResNet-BiLSTM-Attn (no TPS) · ICDAR'13 total edit distanceTED: 260 -> 157Improved
post-tune with LS-EDResNet-BiLSTM-Attn (no TPS) · ICDAR'15accuracy: 71.452 -> 74.655Improved
train the surrogate on random generator data only (global approximation) (ablation)LS-IoU · ICDAR'15 text detectionF1 score: 77.37% -> 74.81%Hurt
train the surrogate on model samples only (local approximation) (ablation)LS-IoU · ICDAR'15 text detectionrecall: 71.21% -> 70.92%Hurt
train the surrogate on model samples only (local approximation) (ablation)LS-IoU · ICDAR'15 text detectionprecision: 84.71% -> 86.60%Improved
use local-global approximation instead of global only (ablation)LS-IoU · ICDAR'15 text detectionF1 score: 74.81% -> 80.66%Improved
use local-global approximation instead of local only (ablation)LS-IoU · ICDAR'15 text detectionF1 score: 77.98% -> 80.66%Improved
replace smooth-L1 regression loss with LS-IoU (local-global) in post-tuningRRPN-ResNet-50 · ICDAR'15 text detectionF1 score: 77.37% -> 80.66%Improved

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