View Extrapolation of Human Body from a Single Image

TL;DR

This paper introduces a novel deep learning pipeline that explicitly estimates human body geometry to synthesize new views from a single image, significantly improving pose variation handling and high-resolution results.

Contribution

It proposes a shape estimation and image generation pipeline with perspective transformation, explicitly leveraging geometry to enhance view extrapolation of articulated human bodies.

Findings

Significantly improved performance on pose-varying human body images.

Effective application to real data captured by 3D sensors.

High-quality, high-resolution view synthesis achieved.

Abstract

We study how to synthesize novel views of human body from a single image. Though recent deep learning based methods work well for rigid objects, they often fail on objects with large articulation, like human bodies. The core step of existing methods is to fit a map from the observable views to novel views by CNNs; however, the rich articulation modes of human body make it rather challenging for CNNs to memorize and interpolate the data well. To address the problem, we propose a novel deep learning based pipeline that explicitly estimates and leverages the geometry of the underlying human body. Our new pipeline is a composition of a shape estimation network and an image generation network, and at the interface a perspective transformation is applied to generate a forward flow for pixel value transportation. Our design is able to factor out the space of data variation and makes learning at each step much easier. Empirically, we show that the performance for pose-varying objects can be improved dramatically. Our method can also be applied on real data captured by 3D sensors, and the flow generated by our methods can be used for generating high quality results in higher resolution.