Deep detection approaches are powerful in controlled conditions, but appear brittle and fail when source models are used off-the-shelf on unseen domains. Most of the existing works on domain adaptation simplify the setting and access jointly both a large source dataset and a sizable amount of target samples. However this scenario is unrealistic in many practical cases as when monitoring image feeds from social media: only a pretrained source model is available and every target image uploaded by the users belongs to a different domain not foreseen during training. We address this challenging setting by presenting an object detection algorithm able to exploit a pre-trained source model and perform unsupervised adaptation by using only one target sample seen at test time. Our multi-task architecture includes a self-supervised branch that we exploit to meta-train the whole model with single-sample cross-domain episodes, and prepare to the test condition. At deployment time the self-supervised task is iteratively solved on any incoming sample to one-shot adapt on it. We introduce a new dataset of social media image feeds and present a thorough benchmark with the most recent cross-domain detection methods showing the advantages of our approach.
Self-supervision & meta-learning for one-shot unsupervised cross-domain detection / Cappio Borlino, Francesco; Polizzotto, Salvatore; Caputo, Barbara; Tommasi, Tatiana. - In: COMPUTER VISION AND IMAGE UNDERSTANDING. - ISSN 1077-3142. - 223:(2022). [10.1016/j.cviu.2022.103549]
Self-supervision & meta-learning for one-shot unsupervised cross-domain detection
Cappio Borlino, Francesco;Caputo, Barbara;Tommasi, Tatiana
2022
Abstract
Deep detection approaches are powerful in controlled conditions, but appear brittle and fail when source models are used off-the-shelf on unseen domains. Most of the existing works on domain adaptation simplify the setting and access jointly both a large source dataset and a sizable amount of target samples. However this scenario is unrealistic in many practical cases as when monitoring image feeds from social media: only a pretrained source model is available and every target image uploaded by the users belongs to a different domain not foreseen during training. We address this challenging setting by presenting an object detection algorithm able to exploit a pre-trained source model and perform unsupervised adaptation by using only one target sample seen at test time. Our multi-task architecture includes a self-supervised branch that we exploit to meta-train the whole model with single-sample cross-domain episodes, and prepare to the test condition. At deployment time the self-supervised task is iteratively solved on any incoming sample to one-shot adapt on it. We introduce a new dataset of social media image feeds and present a thorough benchmark with the most recent cross-domain detection methods showing the advantages of our approach.File | Dimensione | Formato | |
---|---|---|---|
CVIU_oshot_for_publication-1.pdf
non disponibili
Descrizione: Manoscritto accettato per la pubblicazione
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
4.78 MB
Formato
Adobe PDF
|
4.78 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
1-s2.0-S1077314222001278-main.pdf
non disponibili
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
1.88 MB
Formato
Adobe PDF
|
1.88 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2971264