AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Learning Local Descriptors With a CDF-Based Dynamic Soft Margin

Linguang Zhang, Szymon Rusinkiewicz

ICCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 40 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
replace hard-margin triplet loss with CDF-based dynamic soft marginHardNet+ · UBC PhotoTourism mean, real-valuedFPR95 (%): 1.52 -> 1.18Improved
replace hard-margin triplet loss with CDF-based dynamic soft marginL2-Net binary · UBC PhotoTourism mean, binaryFPR95 (%): 3.01 -> 2.63Improved
use dynamic soft margin instead of softplus static soft margin (ablation)triplet loss · Real-valued, UBC FPR95FPR95 (%): 1.20 -> 0.95Improved
use dynamic soft margin instead of Wang et al. dynamic hard margin (ablation)triplet loss · Real-valued, UBC FPR95FPR95 (%): 1.18 -> 0.95Improved
use dynamic soft margin instead of hardest 1/8 mining (ablation)triplet loss · Real-valued, UBC FPR95FPR95 (%): 2.19 -> 0.95Improved
use dynamic soft margin instead of softplus (ablation)triplet loss · Binary, HPatches patch retrievalmAP (%): 54.23 -> 55.12Improved
build the PDF from dpos instead of dpos - dneg (ablation)dynamic soft margin · Real-valued, UBC FPR95FPR95 (%): 0.95 -> 0.98No clear effect
build the PDF from dneg instead of dpos - dneg (ablation)dynamic soft margin · Binary, UBC FPR95FPR95 (%): 2.31 -> 3.02Hurt
approximate PDF with a Gaussian instead of histogram (ablation)dynamic soft margin · Real-valued, UBC FPR95FPR95 (%): 0.95 -> 1.07Hurt
train on union of UBC PhotoTourism and HPatches (Ours++)dynamic soft margin · Oxford5k HQE multiple assignmentmAP (%): 85.94 -> 86.42Improved
tune triplet margin from the default 1.0 to 0.375 (ablation)HardNet · UBC real-valued descriptorFPR95Improved

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