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

Recent developments in monocular multi-object tracking have been verysuccessful in tracking visible objects and bridging short occlusion gaps,mainly relying on data-driven appearance models. While we have significantlyadvanced short-term tracking performance, bridging longer occlusion gapsremains elusive: state-of-the-art object trackers only bridge less than 10% ofocclusions longer than three seconds. We suggest that the missing key isreasoning about future trajectories over a longer time horizon. Intuitively,the longer the occlusion gap, the larger the search space for possibleassociations. In this paper, we show that even a small yet diverse set oftrajectory predictions for moving agents will significantly reduce this searchspace and thus improve long-term tracking robustness. Our experiments suggestthat the crucial components of our approach are reasoning in a bird's-eye viewspace and generating a small yet diverse set of forecasts while accounting fortheir localization uncertainty. This way, we can advance state-of-the-arttrackers on the MOTChallenge dataset and significantly improve their long-termtracking performance. This paper's source code and experimental data areavailable at https://github.com/dendorferpatrick/QuoVadis.

Quo Vadis: Is Trajectory Forecasting the Key Towards Long-Term Multi-Object Tracking?