Total-Recon: Deformable Scene Reconstruction for Embodied View Synthesis

TL;DR

Total-Recon is a novel method that enables photorealistic reconstruction and view synthesis of deformable scenes from long monocular RGBD videos, effectively handling complex articulated motions and scene decomposition.

Contribution

It introduces the first approach to reconstruct deformable scenes from long monocular RGBD videos by hierarchical decomposition of background and objects with articulated motion modeling.

Findings

Outperforms prior methods on challenging videos

Successfully reconstructs complex articulated motions

Enables novel view synthesis from monocular RGBD data

Abstract

We explore the task of embodied view synthesis from monocular videos of deformable scenes. Given a minute-long RGBD video of people interacting with their pets, we render the scene from novel camera trajectories derived from the in-scene motion of actors: (1) egocentric cameras that simulate the point of view of a target actor and (2) 3rd-person cameras that follow the actor. Building such a system requires reconstructing the root-body and articulated motion of every actor, as well as a scene representation that supports free-viewpoint synthesis. Longer videos are more likely to capture the scene from diverse viewpoints (which helps reconstruction) but are also more likely to contain larger motions (which complicates reconstruction). To address these challenges, we present Total-Recon, the first method to photorealistically reconstruct deformable scenes from long monocular RGBD videos. Crucially, to scale to long videos, our method hierarchically decomposes the scene into the background and objects, whose motion is decomposed into carefully initialized root-body motion and local articulations. To quantify such "in-the-wild" reconstruction and view synthesis, we collect ground-truth data from a specialized stereo RGBD capture rig for 11 challenging videos, significantly outperforming prior methods. Our code, model, and data can be found at https://andrewsonga.github.io/totalrecon .