Long-Term Temporally Consistent Unpaired Video Translation from Simulated Surgical 3D Data

TL;DR

This paper introduces a novel method combining unpaired image translation and neural rendering to achieve long-term, view-consistent video translation from simulated to photorealistic surgical scenes, enhancing training and evaluation in medical applications.

Contribution

It proposes a new approach that integrates global textures and view-consistency loss to produce globally consistent, long-term videos from simulated surgical data.

Findings

Achieves long-term temporal consistency in unpaired video translation.

Produces view-consistent, photorealistic surgical videos from simulated data.

Enables better training and evaluation for surgical applications.

Abstract

Research in unpaired video translation has mainly focused on short-term temporal consistency by conditioning on neighboring frames. However for transfer from simulated to photorealistic sequences, available information on the underlying geometry offers potential for achieving global consistency across views. We propose a novel approach which combines unpaired image translation with neural rendering to transfer simulated to photorealistic surgical abdominal scenes. By introducing global learnable textures and a lighting-invariant view-consistency loss, our method produces consistent translations of arbitrary views and thus enables long-term consistent video synthesis. We design and test our model to generate video sequences from minimally-invasive surgical abdominal scenes. Because labeled data is often limited in this domain, photorealistic data where ground truth information from the simulated domain is preserved is especially relevant. By extending existing image-based methods to view-consistent videos, we aim to impact the applicability of simulated training and evaluation environments for surgical applications. Code and data: http://opencas.dkfz.de/video-sim2real.