3D human pose estimation is fundamental to understanding human behavior. Recently, promising results have been achieved by graph convolutional networks (GCNs), which achieve state-of-the-art performance and provide rather light-weight architectures. However, a major limitation of GCNs is their inability to encode all the transformations between joints explicitly. To address this issue, we propose a novel spectral GCN using the Möbius transformation (Möbius-GCN). In particular, this allows us to directly and explicitly encode the transformation between joints, resulting in a significantly more compact representation. Compared to even the lightest architectures so far, our novel approach requires 90–98% fewer parameters, i.e. our lightest MöbiusGCN uses only 0.042M trainable parameters. Besides the drastic parameter reduction, explicitly encoding the transformation of joints also enables us to achieve state-of-the-art results. We evaluate our approach on the two challenging pose estimation benchmarks, Human3.6M and MPI-INF-3DHP, demonstrating both state-of-the-art results and the generalization capabilities of MöbiusGCN.

3D Human Pose Estimation Using Möbius Graph Convolutional Networks / Azizi, N.; Possegger, H.; Rodola, E.; Bischof, H.. - 13661 LNCS:(2022), pp. 160-178. (Intervento presentato al convegno 17th European Conference Computer Vision tenutosi a Tel Aviv, Israel) [10.1007/978-3-031-19769-7_10].

3D Human Pose Estimation Using Möbius Graph Convolutional Networks

Rodola E.;
2022

Abstract

3D human pose estimation is fundamental to understanding human behavior. Recently, promising results have been achieved by graph convolutional networks (GCNs), which achieve state-of-the-art performance and provide rather light-weight architectures. However, a major limitation of GCNs is their inability to encode all the transformations between joints explicitly. To address this issue, we propose a novel spectral GCN using the Möbius transformation (Möbius-GCN). In particular, this allows us to directly and explicitly encode the transformation between joints, resulting in a significantly more compact representation. Compared to even the lightest architectures so far, our novel approach requires 90–98% fewer parameters, i.e. our lightest MöbiusGCN uses only 0.042M trainable parameters. Besides the drastic parameter reduction, explicitly encoding the transformation of joints also enables us to achieve state-of-the-art results. We evaluate our approach on the two challenging pose estimation benchmarks, Human3.6M and MPI-INF-3DHP, demonstrating both state-of-the-art results and the generalization capabilities of MöbiusGCN.
2022
17th European Conference Computer Vision
3D Human Pose; Graph Convolutional Networks; Möbius transformation
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
3D Human Pose Estimation Using Möbius Graph Convolutional Networks / Azizi, N.; Possegger, H.; Rodola, E.; Bischof, H.. - 13661 LNCS:(2022), pp. 160-178. (Intervento presentato al convegno 17th European Conference Computer Vision tenutosi a Tel Aviv, Israel) [10.1007/978-3-031-19769-7_10].
File allegati a questo prodotto
Non ci sono file associati a questo prodotto.

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/11573/1673390
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact