Rethinking matching-based few-shot action recognition

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F23%3A00370588" target="_blank" >RIV/68407700:21230/23:00370588 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-031-31435-3_15" target="_blank" >https://doi.org/10.1007/978-3-031-31435-3_15</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-31435-3_15" target="_blank" >10.1007/978-3-031-31435-3_15</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Rethinking matching-based few-shot action recognition

Popis výsledku v původním jazyce

Few-shot action recognition, i.e. recognizing new action classes given only a few examples, benefits from incorporating temporal information. Prior work either encodes such information in the representation itself and learns classifiers at test time, or obtains frame-level features and performs pairwise temporal matching. We first evaluate a number of matching-based approaches using features from spatio-temporal back- bones, a comparison missing from the literature, and show that the gap in performance between simple baselines and more complicated methods is significantly reduced. Inspired by this, we propose Chamfer++, a non-temporal matching function that achieves state-of-the-art results in few-shot action recognition. We show that, when starting from temporal features, our parameter-free and interpretable approach can outperform all other matching-based and classifier methods for one-shot action recognition on three common datasets without using temporal information in the matching stage. Project page: https://jbertrand89.github.io/matching-based-fsar

Název v anglickém jazyce

Rethinking matching-based few-shot action recognition

Popis výsledku anglicky

Few-shot action recognition, i.e. recognizing new action classes given only a few examples, benefits from incorporating temporal information. Prior work either encodes such information in the representation itself and learns classifiers at test time, or obtains frame-level features and performs pairwise temporal matching. We first evaluate a number of matching-based approaches using features from spatio-temporal back- bones, a comparison missing from the literature, and show that the gap in performance between simple baselines and more complicated methods is significantly reduced. Inspired by this, we propose Chamfer++, a non-temporal matching function that achieves state-of-the-art results in few-shot action recognition. We show that, when starting from temporal features, our parameter-free and interpretable approach can outperform all other matching-based and classifier methods for one-shot action recognition on three common datasets without using temporal information in the matching stage. Project page: https://jbertrand89.github.io/matching-based-fsar

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/GM21-28830M" target="_blank" >GM21-28830M: Učení Univerzální Vizuální Reprezentace s Omezenou Supervizí</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

SCIA 2023: Image Analysis, Part I

ISBN

978-3-031-31434-6

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

22

Strana od-do

215-236

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Sirkka

Datum konání akce

18. 4. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—