EnglishHere we introduce an approximated differentiable renderer to refine a 6-DoF pose prediction using only 2D alignment information. To this end, a two-branched convolutional encoder network is employed to jointly estimate the object class and its 6-DoF pose in the scene. We then propose a new formulation of an approximated differentiable renderer to re-project the 3D object on the image according to its predicted pose; in this way the alignment error between the observed and the re-projected object silhouette can be measured. Since the renderer is differentiable, it is possible to back-propagate through it to correct the estimated pose at test time in an online learning fashion. Eventually we show how to leverage the classification branch to profitably re-project a representative model of the predicted class (i.e. a medoid) instead. Each object in the scene is processed independently and novel viewpoints in which both objects arrangement and mutual pose are preserved can be rendered. Differentiable renderer code is available at:https://github.com/ndrplz/tensorflow-mesh-renderer.
| Data di pubblicazione: | 2018 |
| Titolo: | End-to-end 6-DoF Object Pose Estimation through Differentiable Rasterization |
| Autore/i: | Palazzi, Andrea; Bergamini, Luca; Calderara, Simone; Cucchiara, Rita |
| Autore/i UNIMORE: | |
| Nome del convegno: | Second Workshop on 3D Reconstruction Meets Semantics (3DRMS) |
| Luogo del convegno: | Munich, Germany |
| Data del convegno: | 8 - 14 September 2018 |
| Citazione: | End-to-end 6-DoF Object Pose Estimation through Differentiable Rasterization / Palazzi, Andrea; Bergamini, Luca; Calderara, Simone; Cucchiara, Rita. - (2018). ((Intervento presentato al convegno Second Workshop on 3D Reconstruction Meets Semantics (3DRMS) tenutosi a Munich, Germany nel 8 - 14 September 2018. |
| Tipologia | Relazione in Atti di Convegno |
File in questo prodotto:
| File | Descrizione | Tipologia | |
|---|---|---|---|
| palazzi_eccvw.pdf | Post-print | Open Access Visualizza/Apri |
I documenti presenti in Iris Unimore sono rilasciati con licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia, salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris
Copyright © 2020