Predicting Novel Views Using Generative Adversarial Query Network
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review
Details
| Original language | English |
|---|---|
| Title of host publication | Image Analysis - 21st Scandinavian Conference, SCIA 2019, Proceedings |
| Editors | Michael Felsberg, Per-Erik Forssén, Jonas Unger, Ida-Maria Sintorn |
| Publisher | Springer Verlag |
| Pages | 16-27 |
| Number of pages | 12 |
| ISBN (Print) | 9783030202040 |
| DOIs | |
| Publication status | Published - 2019 |
| Publication type | A4 Article in a conference publication |
| Event | Scandinavian Conference on Image Analysis - Norrköping, Sweden Duration: 11 Jun 2019 → 13 Jun 2019 |
Publication series
| Name | Lecture Notes in Computer Science |
|---|---|
| Volume | 11482 |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Conference
| Conference | Scandinavian Conference on Image Analysis |
|---|---|
| Country | Sweden |
| City | Norrköping |
| Period | 11/06/19 → 13/06/19 |
Abstract
The problem of predicting a novel view of the scene using an arbitrary number of observations is a challenging problem for computers as well as for humans. This paper introduces the Generative Adversarial Query Network (GAQN), a general learning framework for novel view synthesis that combines Generative Query Network (GQN) and Generative Adversarial Networks (GANs). The conventional GQN encodes input views into a latent representation that is used to generate a new view through a recurrent variational decoder. The proposed GAQN builds on this work by adding two novel aspects: First, we extend the current GQN architecture with an adversarial loss function for improving the visual quality and convergence speed. Second, we introduce a feature-matching loss function for stabilizing the training procedure. The experiments demonstrate that GAQN is able to produce high-quality results and faster convergence compared to the conventional approach.
ASJC Scopus subject areas
Keywords
- Generative Adversarial Query Network, Mean feature matching loss, Novel view synthesis