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

Existing deep learning based visual servoing approaches regress the relativecamera pose between a pair of images. Therefore, they require a huge amount oftraining data and sometimes fine-tuning for adaptation to a novel scene.Furthermore, current approaches do not consider underlying geometry of thescene and rely on direct estimation of camera pose. Thus, inaccuracies inprediction of the camera pose, especially for distant goals, lead to adegradation in the servoing performance. In this paper, we propose a two-foldsolution: (i) We consider optical flow as our visual features, which arepredicted using a deep neural network. (ii) These flow features are thensystematically integrated with depth estimates provided by another neuralnetwork using interaction matrix. We further present an extensive benchmark ina photo-realistic 3D simulation across diverse scenes to study the convergenceand generalisation of visual servoing approaches. We show convergence for over3m and 40 degrees while maintaining precise positioning of under 2cm and 1degree on our challenging benchmark where the existing approaches that areunable to converge for majority of scenarios for over 1.5m and 20 degrees.Furthermore, we also evaluate our approach for a real scenario on an aerialrobot. Our approach generalizes to novel scenarios producing precise and robustservoing performance for 6 degrees of freedom positioning tasks with even largecamera transformations without any retraining or fine-tuning.

DFVS: Deep Flow Guided Scene Agnostic Image Based Visual Servoing