Unsupervised Domain Adaptation (UDA) for point cloud classification is an emerging research problem with relevant practical motivations. Reliance on multi-task learning to align features across domains has been the standard way to tackle it. In this paper, we take a different path and propose RefRec, the first approach to investigate pseudo-labels and self-training in UDA for point clouds. We present two main innovations to make self-training effective on 3D data: i) refinement of noisy pseudo-labels by matching shape descriptors that are learned by the unsupervised task of shape reconstruction on both domains; ii) a novel self-training protocol that learns domain-specific decision boundaries and reduces the negative impact of mislabelled target samples and in-domain intra-class variability. RefRec sets the new state of the art in both standard benchmarks used to test UDA for point cloud classification, showcasing the effectiveness of self-training for this important problem.

RefRec: Pseudo-labels Refinement via Shape Reconstruction for Unsupervised 3D Domain Adaptation / Adriano Cardace; Riccardo Spezialetti ; Pierluigi Zama Ramirez; Samuele Salti; Luigi Di Stefano. - ELETTRONICO. - (2021), pp. 331-341. (Intervento presentato al convegno 9th International Conference on 3D Vision (3DV) tenutosi a London, United Kingdom nel Dec. 1 2021 to Dec. 3 2021) [10.1109/3DV53792.2021.00043].

RefRec: Pseudo-labels Refinement via Shape Reconstruction for Unsupervised 3D Domain Adaptation

Adriano Cardace
;
Riccardo Spezialetti;Pierluigi Zama Ramirez;Samuele Salti;Luigi Di Stefano
2021

Abstract

Unsupervised Domain Adaptation (UDA) for point cloud classification is an emerging research problem with relevant practical motivations. Reliance on multi-task learning to align features across domains has been the standard way to tackle it. In this paper, we take a different path and propose RefRec, the first approach to investigate pseudo-labels and self-training in UDA for point clouds. We present two main innovations to make self-training effective on 3D data: i) refinement of noisy pseudo-labels by matching shape descriptors that are learned by the unsupervised task of shape reconstruction on both domains; ii) a novel self-training protocol that learns domain-specific decision boundaries and reduces the negative impact of mislabelled target samples and in-domain intra-class variability. RefRec sets the new state of the art in both standard benchmarks used to test UDA for point cloud classification, showcasing the effectiveness of self-training for this important problem.
2021
2021 International Conference on 3D Vision (3DV)
331
341
RefRec: Pseudo-labels Refinement via Shape Reconstruction for Unsupervised 3D Domain Adaptation / Adriano Cardace; Riccardo Spezialetti ; Pierluigi Zama Ramirez; Samuele Salti; Luigi Di Stefano. - ELETTRONICO. - (2021), pp. 331-341. (Intervento presentato al convegno 9th International Conference on 3D Vision (3DV) tenutosi a London, United Kingdom nel Dec. 1 2021 to Dec. 3 2021) [10.1109/3DV53792.2021.00043].
Adriano Cardace; Riccardo Spezialetti ; Pierluigi Zama Ramirez; Samuele Salti; Luigi Di Stefano
File in questo prodotto:
File Dimensione Formato  
main.pdf

accesso aperto

Tipo: Postprint
Licenza: Licenza per accesso libero gratuito
Dimensione 764.13 kB
Formato Adobe PDF
764.13 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/11585/864997
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 8
  • ???jsp.display-item.citation.isi??? 4
social impact