AI Paper Atlas

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

Matching neural paths: transfer from recognition to correspondence search

Nikolay Savinov, Lubor Ladicky, Marc Pollefeys

NeurIPS 2017 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 8 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
use hierarchical path matching instead of stacked-feature correlationVGG-16 transfer for stereo correspondence · KITTI 2012 validationErr_3 (%): 5.23 -> 3.94Improved
use hierarchical path matching instead of stacked-feature correlationVGG-16 transfer for stereo correspondence · KITTI 2015 validationErr_3 (%): 6.16 -> 4.78Improved
use hierarchical path matching instead of the best listed learning-based baseline, acrtVGG-16 transfer for stereo correspondence · KITTI 2012 validationErr_3 (%): 2.61 -> 3.94Hurt
use hierarchical path matching instead of the best listed learning-based baseline, acrtVGG-16 transfer for stereo correspondence · KITTI 2015 validationErr_3 (%): 3.25 -> 4.78Hurt
add the full feature hierarchy rather than using only layer 2 (ablation)ours(2, 8) · KITTI 2012 validationErr_3 (%): 5.28 -> 3.94Improved
add the full feature hierarchy rather than using only layers 2 through 3 (ablation)ours(2, 8) · KITTI 2012 validationErr_3 (%): 5.41 -> 3.94Improved
use full convolutional receptive-field coverage rather than only central arcs (ablation)ours(2, 8) · KITTI 2012 validationErr_3 (%): 4.02 -> 3.94Improved
use sum/product aggregation rather than the alternative max-based aggregation (ablation)neural-path matching · stereo correspondenceErr_3 (%)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