A Shape-Aware Retargeting Approach to Transfer Human Motion and Appearance in Monocular Videos

TL;DR

This paper introduces a shape-aware hybrid rendering method for human motion and appearance transfer in monocular videos, outperforming many existing neural approaches and providing a new challenging benchmark dataset.

Contribution

A novel shape-aware retargeting approach that incorporates physical constraints and a new benchmark dataset for evaluating video retargeting methods.

Findings

Competitive visual quality against state-of-the-art neural methods

Effective transfer across different shapes, motions, and camera setups

Public availability of dataset and code for community use

Abstract

Transferring human motion and appearance between videos of human actors remains one of the key challenges in Computer Vision. Despite the advances from recent image-to-image translation approaches, there are several transferring contexts where most end-to-end learning-based retargeting methods still perform poorly. Transferring human appearance from one actor to another is only ensured when a strict setup has been complied, which is generally built considering their training regime's specificities. In this work, we propose a shape-aware approach based on a hybrid image-based rendering technique that exhibits competitive visual retargeting quality compared to state-of-the-art neural rendering approaches. The formulation leverages the user body shape into the retargeting while considering physical constraints of the motion in 3D and the 2D image domain. We also present a new video retargeting benchmark dataset composed of different videos with annotated human motions to evaluate the task of synthesizing people's videos, which can be used as a common base to improve tracking the progress in the field. The dataset and its evaluation protocols are designed to evaluate retargeting methods in more general and challenging conditions. Our method is validated in several experiments, comprising publicly available videos of actors with different shapes, motion types, and camera setups. The dataset and retargeting code are publicly available to the community at: https://www.verlab.dcc.ufmg.br/retargeting-motion.