Consistent Depth of Moving Objects in Video

TL;DR

This paper introduces a novel test-time training framework that estimates consistent depth in videos with moving objects and camera motion, enabling realistic 3D scene understanding and editing effects.

Contribution

It proposes a new method combining depth prediction CNN and scene-flow MLP trained jointly for temporally consistent depth estimation in dynamic scenes.

Findings

Achieves accurate, temporally coherent depth maps on diverse videos.

Enables depth-and-motion aware video editing effects.

Demonstrates robustness to various moving objects and camera motions.

Abstract

We present a method to estimate depth of a dynamic scene, containing arbitrary moving objects, from an ordinary video captured with a moving camera. We seek a geometrically and temporally consistent solution to this underconstrained problem: the depth predictions of corresponding points across frames should induce plausible, smooth motion in 3D. We formulate this objective in a new test-time training framework where a depth-prediction CNN is trained in tandem with an auxiliary scene-flow prediction MLP over the entire input video. By recursively unrolling the scene-flow prediction MLP over varying time steps, we compute both short-range scene flow to impose local smooth motion priors directly in 3D, and long-range scene flow to impose multi-view consistency constraints with wide baselines. We demonstrate accurate and temporally coherent results on a variety of challenging videos containing diverse moving objects (pets, people, cars), as well as camera motion. Our depth maps give rise to a number of depth-and-motion aware video editing effects such as object and lighting insertion.