Data-Efficient Ranking Distillation for Image Retrieval

Zakaria Laskar*, Juho Kannala

*Tämän työn vastaava kirjoittaja

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

76 Lataukset (Pure)

Abstrakti

Recent advances in deep learning has lead to rapid developments in the field of image retrieval. However, the best performing architectures incur significant computational cost. The paper addresses this issue using knowledge distillation for metric learning problems. Unlike previous approaches, our proposed method jointly addresses the following constraints: i) limited queries to teacher model, ii) black box teacher model with access to the final output representation, and iii) small fraction of original training data without any ground-truth labels. In addition, the distillation method does not require the student and teacher to have same dimensionality. The key idea is to augment the original training set with additional samples by performing linear interpolation in the final output representation space. In low training sample settings, our approach outperforms the fully supervised baseline approach on ROxford5k and RParis6k with the least possible teacher supervision.

AlkuperäiskieliEnglanti
OtsikkoComputer Vision – ACCV 2020 - 15th Asian Conference on Computer Vision, 2020, Revised Selected Papers
ToimittajatHiroshi Ishikawa, Cheng-Lin Liu, Tomas Pajdla, Jianbo Shi
KustantajaSpringer
Sivut469-484
Sivumäärä16
ISBN (painettu)9783030695248
DOI - pysyväislinkit
TilaJulkaistu - 2021
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaAsian Conference on Computer Vision - Virtual, Online
Kesto: 30 marrask. 20204 jouluk. 2020
Konferenssinumero: 15

Julkaisusarja

NimiLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vuosikerta12622 LNCS
ISSN (painettu)0302-9743
ISSN (elektroninen)1611-3349

Conference

ConferenceAsian Conference on Computer Vision
LyhennettäACCV
KaupunkiVirtual, Online
Ajanjakso30/11/202004/12/2020

Sormenjälki

Sukella tutkimusaiheisiin 'Data-Efficient Ranking Distillation for Image Retrieval'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä