In the last few years large-scale foundational models have shown remarkable performance in computer vision tasks. However, deploying such models in a production environment poses a significant challenge, because of their computational requirements. Furthermore, these models typically produce generic results and they often need some sort of external input. The concept of knowledge distillation provides a promising solution to this problem. In this paper, we focus on the challenges faced in the application of knowledge distillation techniques in the task of augmenting a dataset for object detection used in a commercial Visual Recommender System called VISIDEA; the goal consists in detecting items in various e-commerce websites, encompassing a wide range of custom product categories. We discuss a possible solution to problems such as label duplication, erroneous labeling and lack of robustness to prompting, by considering examples in the field of fashion apparel recommendation.

Knowledge Distillation for a Domain-Adaptive Visual Recommender System

Luigi Portinale
2024-01-01

Abstract

In the last few years large-scale foundational models have shown remarkable performance in computer vision tasks. However, deploying such models in a production environment poses a significant challenge, because of their computational requirements. Furthermore, these models typically produce generic results and they often need some sort of external input. The concept of knowledge distillation provides a promising solution to this problem. In this paper, we focus on the challenges faced in the application of knowledge distillation techniques in the task of augmenting a dataset for object detection used in a commercial Visual Recommender System called VISIDEA; the goal consists in detecting items in various e-commerce websites, encompassing a wide range of custom product categories. We discuss a possible solution to problems such as label duplication, erroneous labeling and lack of robustness to prompting, by considering examples in the field of fashion apparel recommendation.
File in questo prodotto:
File Dimensione Formato  
FLAIRS_37_51Camera_ready.pdf

file ad accesso aperto

Descrizione: paper proceedings
Tipologia: Versione Editoriale (PDF)
Licenza: Creative commons
Dimensione 199.14 kB
Formato Adobe PDF
199.14 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11579/179082
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact