Salvato in:
Dettagli Bibliografici
Autori principali: Peirone, Simone Alberto, Goletto, Gabriele, Planamente, Mirco, Bottino, Andrea, Caputo, Barbara, Averta, Giuseppe
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2409.14205
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912039760297984
author Peirone, Simone Alberto
Goletto, Gabriele
Planamente, Mirco
Bottino, Andrea
Caputo, Barbara
Averta, Giuseppe
author_facet Peirone, Simone Alberto
Goletto, Gabriele
Planamente, Mirco
Bottino, Andrea
Caputo, Barbara
Averta, Giuseppe
contents Human activities exhibit a strong correlation between actions and the places where these are performed, such as washing something at a sink. More specifically, in daily living environments we may identify particular locations, hereinafter named activity-centric zones, which may afford a set of homogeneous actions. Their knowledge can serve as a prior to favor vision models to recognize human activities. However, the appearance of these zones is scene-specific, limiting the transferability of this prior information to unfamiliar areas and domains. This problem is particularly relevant in egocentric vision, where the environment takes up most of the image, making it even more difficult to separate the action from the context. In this paper, we discuss the importance of decoupling the domain-specific appearance of activity-centric zones from their universal, domain-agnostic representations, and show how the latter can improve the cross-domain transferability of Egocentric Action Recognition (EAR) models. We validate our solution on the EPIC-Kitchens-100 and Argo1M datasets
format Preprint
id arxiv_https___arxiv_org_abs_2409_14205
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Egocentric zone-aware action recognition across environments
Peirone, Simone Alberto
Goletto, Gabriele
Planamente, Mirco
Bottino, Andrea
Caputo, Barbara
Averta, Giuseppe
Computer Vision and Pattern Recognition
Human activities exhibit a strong correlation between actions and the places where these are performed, such as washing something at a sink. More specifically, in daily living environments we may identify particular locations, hereinafter named activity-centric zones, which may afford a set of homogeneous actions. Their knowledge can serve as a prior to favor vision models to recognize human activities. However, the appearance of these zones is scene-specific, limiting the transferability of this prior information to unfamiliar areas and domains. This problem is particularly relevant in egocentric vision, where the environment takes up most of the image, making it even more difficult to separate the action from the context. In this paper, we discuss the importance of decoupling the domain-specific appearance of activity-centric zones from their universal, domain-agnostic representations, and show how the latter can improve the cross-domain transferability of Egocentric Action Recognition (EAR) models. We validate our solution on the EPIC-Kitchens-100 and Argo1M datasets
title Egocentric zone-aware action recognition across environments
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.14205