AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Object-centered image stitching

Charles Herrmann, Chen Wang, Richard Strong Bowen, Emil Keyder, Ramin Zabih

ECCV 2018 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning · 38 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
add the cropping, duplication and occlusion terms to the seam-finding MRF energy (ablation)object-centered seam finding · difficult stitching image sets (ablation of pairwise terms)stitching quality (visual)Improved
select among multiple registrations by lowest final energy and best evaluation score instead of a single registration (ablation)registration stage · difficult stitching image setsstitching qualityNo clear effect
use the object-centered MRF terms instead of Google Photos and APAPstitching pipeline · Stitching example (humans, Fig. 4)max MS-SSIM: 0.1140 -> 0.4882Improved
use the object-centered MRF terms instead of NIS and APAPstitching pipeline · Stitching example (bikes, Fig. 5)max MS-SSIM: 0.3971 -> 0.8965Improved
use the object-centered MRF terms instead of Photoshopstitching pipeline · Stitching example (cats, Fig. 6)max MS-SSIM: 0.5099 -> 0.6769Improved
add object-based terms to a scene with objects that are not detectable (mountains, rivers)object-centered seam finding · natural scenes without detectable objectsstitching qualityNo clear effect

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