cVIL: Class-Centric Visual Interactive Labeling

TL;DR

cVIL introduces a class-centric visual interactive labeling method that enhances annotation efficiency and accuracy for large image datasets through property measures, batch labeling, and user-friendly interactions.

Contribution

The paper proposes cVIL, a novel class-centric approach that improves visual interactive labeling by supporting instance and batch labeling with property measures.

Findings

cVIL with batch labeling outperforms traditional active learning methods.

User study shows cVIL achieves better accuracy and user preference.

Simulated experiments validate cVIL's effectiveness in complex data annotation.

Abstract

We present cVIL, a class-centric approach to visual interactive labeling, which facilitates human annotation of large and complex image data sets. cVIL uses different property measures to support instance labeling for labeling difficult instances and batch labeling to quickly label easy instances. Simulated experiments reveal that cVIL with batch labeling can outperform traditional labeling approaches based on active learning. In a user study, cVIL led to better accuracy and higher user preference compared to a traditional instance-based visual interactive labeling approach based on 2D scatterplots.