Bag of Bags uses per-fragment k-means vocabularies from autoencoder embeddings to improve manuscript join retrieval, reaching Hit@1 of 0.78 versus 0.74 for the best BoW baseline on Genizah data.
Identifying join candidates in the Cairo Genizah.Interna- tional Journal of Computer Vision, 94(1):118–135
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bag of Bags: Adaptive Visual Vocabularies for Genizah Join Image Retrieval
Bag of Bags uses per-fragment k-means vocabularies from autoencoder embeddings to improve manuscript join retrieval, reaching Hit@1 of 0.78 versus 0.74 for the best BoW baseline on Genizah data.