AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Localizing and Orienting Street Views Using Overhead Imagery

Nam N. Vo, James Hays

ECCV 2016 · Image generation › Autoregressive and latent generation · approach: Retrieval and matching with generative features · 310 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 contrastive loss with distance-based logistic (DBL) layer in Siamese network (ablation)Siamese network · Denver test setranking recall @top 1%: 21.6 -> 48.4Improved
replace hinge triplet loss with DBL layer (ablation)Triplet network · Denver test setranking recall @top 1%: 43.2 -> 49.3Improved
use triplet instead of Siamese contrastive network (ablation)cross-view embedding · Denver test setranking recall @top 1%: 21.6 -> 43.2Improved
process both images jointly in hybrid classification network instead of pure classification network (ablation)classification network · Denver test setclassification accuracy: 90.0 -> 91.5Improved
increase rotation invariance from 0 to 90 degrees with 16 test rotations (ablation)triplet DBL-Net · Denver test setranking recall @top 1%: 37.3 -> 49.3Improved
full 360 degree rotation invariance instead of 90 degrees at 16 test rotations (ablation)triplet DBL-Net · Denver test setranking recall @top 1%: 49.3 -> 41.9Hurt
add orientation regression to 360 degree RI network (ablation)triplet DBL-Net · Denver test setranking recall @top 1%: 41.9 -> 54.8Improved
add orientation regression to 90 degree RI network (ablation)triplet DBL-Net · Denver test setranking recall @top 1%: 49.3 -> 47.4Hurt
increase test rotation samples from 1 to 16 for 0 degree RI network (ablation)triplet DBL-Net · Denver test setranking recall @top 1%: 11.0 -> 37.3Improved
average features of 16 rotation samples at indexing time instead of testing all 16 (ablation)360 RI + OR network · Denver test setranking recall @top 1%: 54.8 -> 54.0No clear effect
use all triplets in mini-batch (eDBL) instead of single triplets (ablation)triplet DBL-Net 360 RI + OR + avg16 · Denver test setranking recall @top 1%: 54.0 -> 59.9Improved

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