AI Paper Atlas

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

Cross-Descriptor Visual Localization and Mapping

Mihai Dusmanu, Ondrej Miksik, Johannes L. Schönberger, Marc Pollefeys

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 32 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
translate SIFT map descriptors to HardNet instead of keeping SIFTSIFT map · Aachen day, 0.5m 5deg% localized queries: 88.7 -> 90.9Improved
translate BRIEF map descriptors to HardNet instead of keeping BRIEFBRIEF map · Aachen night, 0.5m 5deg% localized queries: 36.7 -> 49.0Improved
translate SIFT map to BRIEF queries instead of native SIFTSIFT map · Aachen night, 0.5m 5deg% localized queries: 61.2 -> 23.5Hurt
translate HardNet descriptors to SIFT/BRIEF in continuous deploymentHardNet map · Aachen night, 0.5m 5deg% localized queries: 72.4 -> 26.5Hurt
use the joint embedding for heterogeneous map instead of native BRIEF queriesBRIEF queries · Aachen day, 0.5m 5deg% localized queries: 81.4 -> 86.7Improved
use joint-embedding collaborative map instead of native HardNetHardNet · Aachen night, 0.5m 5deg% localized queries: 72.4 -> 61.2Hurt
use Embed instead of per-quarter native SOSNet mapscollaborative mapping · Madrid Metropolis, infinite threshold% localized images: 21.2 -> 92.1Improved
use Embed instead of Progressive translationcollaborative mapping · Tower of London, infinite threshold% localized images: 96.6 -> 97.9Improved
increase joint embedding dimension from 128 to 256 (ablation)Embed · Aachen night, 0.5m 5deg% localized queries: 60.2 -> 64.3No clear effect
reduce joint embedding dimension from 128 to 32 (ablation)Embed · Aachen night, 0.5m 5deg% localized queries: 60.2 -> 49.0Hurt
reduce joint embedding dimension from 128 to 16 (ablation)Embed · Aachen night, 0.5m 5deg% localized queries: 60.2 -> 33.7Hurt
add matching loss to the encoder-decoder (ablation)encoder-decoder · HPatches sequencesmatching accuracyImproved

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