English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Estimating precise metric depth and scene reconstruction from monocularendoscopy is a fundamental task for surgical navigation in robotic surgery.However, traditional stereo matching adopts binocular images to perceive thedepth information, which is difficult to transfer to the soft robotics-basedsurgical systems due to the use of monocular endoscopy. In this paper, wepresent a novel framework that combines robot kinematics and monocularendoscope images with deep unsupervised learning into a single network formetric depth estimation and then achieve 3D reconstruction of complex anatomy.Specifically, we first obtain the relative depth maps of surgical scenes byleveraging a brightness-aware monocular depth estimation method. Then, thecorresponding endoscope poses are computed based on non-linear optimization ofgeometric and photometric reprojection residuals. Afterwards, we develop aDepth-driven Sliding Optimization (DDSO) algorithm to extract the scalingcoefficient from kinematics and calculated poses offline. By coupling themetric scale and relative depth data, we form a robust ensemble that representsthe metric and consistent depth. Next, we treat the ensemble as supervisorylabels to train a metric depth estimation network for surgeries (i.e.,MetricDepthS-Net) that distills the embeddings from the robot kinematics,endoscopic videos, and poses. With accurate metric depth estimation, we utilizea dense visual reconstruction method to recover the 3D structure of the wholesurgical site. We have extensively evaluated the proposed framework on publicSCARED and achieved comparable performance with stereo-based depth estimationmethods. Our results demonstrate the feasibility of the proposed approach torecover the metric depth and 3D structure with monocular inputs.

Distilled Visual and Robot Kinematics Embeddings for Metric Depth Estimation in Monocular Scene Reconstruction