DyST: Towards Dynamic Neural Scene Representations on Real-World Videos

Maximilian Seitzer, Sjoerd van Steenkiste, Thomas Kipf, Klaus Greff, Mehdi S. M. Sajjadi

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: neural scene representations, scene representations, representation learning, novel view synthesis

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We introduce the Dynamic Scene Transformer, a model which learns scene representations from real videos and achieves separate control over camera and dynamics.

Abstract: Visual understanding of the world goes beyond the semantics and flat structure of individual images. In this work, we aim to capture both the 3D structure and dynamics of real-world scenes from monocular real-world videos. Our Dynamic Scene Transformer (DyST) model leverages recent work in neural scene representation to learn a latent decomposition of monocular real-world videos into scene content, per-view scene dynamics, and camera pose. This separation is achieved through a novel co-training scheme on monocular videos and our new synthetic dataset DySO. DyST learns tangible latent representations for dynamic scenes that enable view generation with separate control over the camera and the content of the scene.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Primary Area: representation learning for computer vision, audio, language, and other modalities

Submission Number: 5714