Benchmarking the Domain Gap

Benchmarking the Domain Gap measures how model rankings change across video capsule endoscopy datasets and shared-label targets.

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorydomain
Subcategoryvideo capsule endoscopy
Metricclassification performance and ranking stability
Directionhigher_is_better
Unitscore
PublisherDan Hanson and Debesh Jha

What it measures

The study evaluates image-classification backbones trained on Kvasir-Capsule and tested across Kvasir-Capsule, Capsule Vision 2024, and a shared-label subset of Galar. It measures cross-target ranking stability under domain shift.

Task format

Supervised image classification on endoscopy frames or clips with shared-label evaluation.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub