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

We consider the task of re-calibrating the 3D pose of a static surveillancecamera, whose pose may change due to external forces, such as birds, wind,falling objects or earthquakes. Conventionally, camera pose estimation can besolved with a PnP (Perspective-n-Point) method using 2D-to-3D featurecorrespondences, when 3D points are known. However, 3D point annotations arenot always available or practical to obtain in real-world applications. Wepropose an alternative strategy for extracting 3D information to solve forcamera pose by using pedestrian trajectories. We observe that 2D pedestriantrajectories indirectly contain useful 3D information that can be used forinferring camera pose. To leverage this information, we propose a data-drivenapproach by training a neural network (NN) regressor to model a direct mappingfrom 2D pedestrian trajectories projected on the image plane to 3D camera pose.We demonstrate that our regressor trained only on synthetic data can bedirectly applied to real data, thus eliminating the need to label any realdata. We evaluate our method across six different scenes from the Town CentreStreet and DUKEMTMC datasets. Our method achieves an improvement of $\sim50\%$on both position and orientation prediction accuracy when compared to otherSOTA methods.

Estimating 3D Camera Pose from 2D Pedestrian Trajectories