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

Locating 3D objects from a single RGB image via Perspective-n-Points (PnP) isa long-standing problem in computer vision. Driven by end-to-end deep learning,recent studies suggest interpreting PnP as a differentiable layer, so that2D-3D point correspondences can be partly learned by backpropagating thegradient w.r.t. object pose. Yet, learning the entire set of unrestricted 2D-3Dpoints from scratch fails to converge with existing approaches, since thedeterministic pose is inherently non-differentiable. In this paper, we proposethe EPro-PnP, a probabilistic PnP layer for general end-to-end pose estimation,which outputs a distribution of pose on the SE(3) manifold, essentiallybringing categorical Softmax to the continuous domain. The 2D-3D coordinatesand corresponding weights are treated as intermediate variables learned byminimizing the KL divergence between the predicted and target posedistribution. The underlying principle unifies the existing approaches andresembles the attention mechanism. EPro-PnP significantly outperformscompetitive baselines, closing the gap between PnP-based method and thetask-specific leaders on the LineMOD 6DoF pose estimation and nuScenes 3Dobject detection benchmarks.

EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation