Robust Pose Transfer with Dynamic Details using Neural Video Rendering

TL;DR

This paper introduces a neural video rendering framework with a novel texture representation and temporal loss, enabling high-quality pose transfer with dynamic details from short monocular videos, outperforming existing methods.

Contribution

It proposes a new neural rendering approach combining explicit 3D features and learned components, with a texture representation and temporal loss to improve detail preservation and stability.

Findings

Achieves clearer dynamic details in pose transfer videos.

Performs robustly on short videos with only 2k-4k frames.

Outperforms existing methods in detail quality and stability.

Abstract

Pose transfer of human videos aims to generate a high fidelity video of a target person imitating actions of a source person. A few studies have made great progress either through image translation with deep latent features or neural rendering with explicit 3D features. However, both of them rely on large amounts of training data to generate realistic results, and the performance degrades on more accessible internet videos due to insufficient training frames. In this paper, we demonstrate that the dynamic details can be preserved even trained from short monocular videos. Overall, we propose a neural video rendering framework coupled with an image-translation-based dynamic details generation network (D2G-Net), which fully utilizes both the stability of explicit 3D features and the capacity of learning components. To be specific, a novel texture representation is presented to encode both the static and pose-varying appearance characteristics, which is then mapped to the image space and rendered as a detail-rich frame in the neural rendering stage. Moreover, we introduce a concise temporal loss in the training stage to suppress the detail flickering that is made more visible due to high-quality dynamic details generated by our method. Through extensive comparisons, we demonstrate that our neural human video renderer is capable of achieving both clearer dynamic details and more robust performance even on accessible short videos with only 2k - 4k frames.