Content-aware Warping for View Synthesis

TL;DR

This paper introduces content-aware warping, a neural network-based approach that adaptively learns pixel interpolation weights for improved view synthesis, effectively handling occlusions and spatial correlations, outperforming existing methods.

Contribution

It proposes a novel content-aware warping module with an end-to-end framework for view synthesis, incorporating confidence blending, spatial refinement, and a regularization loss.

Findings

Significantly outperforms state-of-the-art methods quantitatively.

Achieves superior visual quality in synthesized views.

Effective handling of occlusions and spatial correlations.

Abstract

Existing image-based rendering methods usually adopt depth-based image warping operation to synthesize novel views. In this paper, we reason the essential limitations of the traditional warping operation to be the limited neighborhood and only distance-based interpolation weights. To this end, we propose content-aware warping, which adaptively learns the interpolation weights for pixels of a relatively large neighborhood from their contextual information via a lightweight neural network. Based on this learnable warping module, we propose a new end-to-end learning-based framework for novel view synthesis from a set of input source views, in which two additional modules, namely confidence-based blending and feature-assistant spatial refinement, are naturally proposed to handle the occlusion issue and capture the spatial correlation among pixels of the synthesized view, respectively. Besides, we also propose a weight-smoothness loss term to regularize the network. Experimental results on light field datasets with wide baselines and multi-view datasets show that the proposed method significantly outperforms state-of-the-art methods both quantitatively and visually. The source code will be publicly available at https://github.com/MantangGuo/CW4VS.