Meta-Personalizing Vision-Language Models to Find Named Instances in Video

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F23%3A00372042" target="_blank" >RIV/68407700:21730/23:00372042 - isvavai.cz</a>

Result on the web

<a href="https://doi.org/10.1109/CVPR52729.2023.01833" target="_blank" >https://doi.org/10.1109/CVPR52729.2023.01833</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/CVPR52729.2023.01833" target="_blank" >10.1109/CVPR52729.2023.01833</a>

Result language

angličtina

Original language name

Meta-Personalizing Vision-Language Models to Find Named Instances in Video

Original language description

Large-scale vision-language models (VLM) have shown impressive results for language-guided search applications. While these models allow category-level queries, they currently struggle with personalized searches for moments in a video where a specific object instance such as “My dog Biscuit” appears. We present the following three contributions to address this problem. First, we describe a method to meta-personalize a pre-trained VLM, i.e., learning how to learn to personalize a VLM at test time to search in video. Our method extends the VLM's token vocabulary by learning novel word embeddings specific to each instance. To capture only instance-specific features, we represent each instance embedding as a combination of shared and learned global category features. Second, we propose to learn such personalization without explicit human supervision. Our approach automatically identifies moments of named visual instances in video using transcripts and vision-language similarity in the VLM's embedding space. Finally, we introduce This-Is-My, a personal video instance retrieval benchmark. We evaluate our approach on This-Is-My and Deep-Fashion2 and show that we obtain a 15% relative improvement over the state of the art on the latter dataset.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year

2023

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

ISBN

979-8-3503-0130-4

ISSN

1063-6919

e-ISSN

2575-7075

Number of pages

10

Pages from-to

19123-19132

Publisher name

IEEE Computer Society

Place of publication

USA

Event location

Vancouver

Event date

Jun 18, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

001062531303042