In this paper we address the problem of generating person images conditioned on a given pose. Specifically, given an image of a person and a target pose, we synthesize a new image of that person in the novel pose. In order to deal with pixel-to-pixel misalignments caused by the pose differences, we introduce deformable skip connections in the generator of our Generative Adversarial Network. Moreover, a nearest-neighbour loss is proposed instead of the common L 1 and L 2 losses in order to match the details of the generated image with the target image. We test our approach using photos of persons in different poses and we compare our method with previous work in this area showing state-of-the-art results in two benchmarks. Our method can be applied to the wider field of deformable object generation, provided that the pose of the articulated object can be extracted using a keypoint detector.

Deformable GANs for Pose-Based Human Image Generation / Siarohin, Aliaksandr; Sangineto, Enver; Lathuiliere, Stephane; Sebe, Nicu. - (2018), pp. 3408-3416. (Intervento presentato al convegno CVPR tenutosi a Salt Lake City nel 18-23 June 2018) [10.1109/CVPR.2018.00359].

Deformable GANs for Pose-Based Human Image Generation

Siarohin, Aliaksandr;Sangineto, Enver;Lathuiliere, Stephane;Sebe, Nicu
2018-01-01

Abstract

In this paper we address the problem of generating person images conditioned on a given pose. Specifically, given an image of a person and a target pose, we synthesize a new image of that person in the novel pose. In order to deal with pixel-to-pixel misalignments caused by the pose differences, we introduce deformable skip connections in the generator of our Generative Adversarial Network. Moreover, a nearest-neighbour loss is proposed instead of the common L 1 and L 2 losses in order to match the details of the generated image with the target image. We test our approach using photos of persons in different poses and we compare our method with previous work in this area showing state-of-the-art results in two benchmarks. Our method can be applied to the wider field of deformable object generation, provided that the pose of the articulated object can be extracted using a keypoint detector.
2018
IEEE/CVF Conference on Computer Vision and Pattern Recognition
New York
IEEE
978-1-5386-6420-9
Siarohin, Aliaksandr; Sangineto, Enver; Lathuiliere, Stephane; Sebe, Nicu
Deformable GANs for Pose-Based Human Image Generation / Siarohin, Aliaksandr; Sangineto, Enver; Lathuiliere, Stephane; Sebe, Nicu. - (2018), pp. 3408-3416. (Intervento presentato al convegno CVPR tenutosi a Salt Lake City nel 18-23 June 2018) [10.1109/CVPR.2018.00359].
File in 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/11572/226661
 Attenzione

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

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 361
  • ???jsp.display-item.citation.isi??? 282
social impact