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 a dense depth map from a single view is geometrically ill-posed,and state-of-the-art methods rely on learning depth's relation with visualappearance using deep neural networks. On the other hand, Structure from Motion(SfM) leverages multi-view constraints to produce very accurate but sparsemaps, as matching across images is typically limited by locally discriminativetexture. In this work, we combine the strengths of both approaches by proposinga novel test-time refinement (TTR) method, denoted as SfM-TTR, that boosts theperformance of single-view depth networks at test time using SfM multi-viewcues. Specifically, and differently from the state of the art, we use sparseSfM point clouds as test-time self-supervisory signal, fine-tuning the networkencoder to learn a better representation of the test scene. Our results showhow the addition of SfM-TTR to several state-of-the-art self-supervised andsupervised networks improves significantly their performance, outperformingprevious TTR baselines mainly based on photometric multi-view consistency. Thecode is available at https://github.com/serizba/SfM-TTR.

SfM-TTR: Using Structure from Motion for Test-Time Refinement of Single-View Depth Networks