← Back to literature explorer
Open-World / Open-Vocabulary From survey PDF — verify links

Anomize

Anomize — Novel-Category Alignment

CVPR · 2025

Survey summary

Anomize targets two challenges in open-vocabulary VAD -- detection ambiguity and categorization confusion -- using multi-level visual data with matching textual information to score novel anomalies, plus label-relation-guided text encoding that better aligns novel videos with their labels.

Open-World

Open-WorldOpen-VocabularyNovel-Category Alignment
Open-World / Open-VocabularyDual-streamGroup-guided text encoding