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

Trajectory forecasting is critical for autonomous platforms to make safeplanning and actions. Currently, most trajectory forecasting methods assumethat object trajectories have been extracted and directly develop trajectorypredictors based on the ground truth trajectories. However, this assumptiondoes not hold in practical situations. Trajectories obtained from objectdetection and tracking are inevitably noisy, which could cause seriousforecasting errors to predictors built on ground truth trajectories. In thispaper, we propose a trajectory predictor directly based on detection resultswithout relying on explicitly formed trajectories. Different from thetraditional methods which encode the motion cue of an agent based on itsclearly defined trajectory, we extract the motion information only based on theaffinity cues among detection results, in which an affinity-aware state updatemechanism is designed to take the uncertainty of association into account. Inaddition, considering that there could be multiple plausible matchingcandidates, we aggregate the states of them. This design relaxes theundesirable effect of noisy trajectory obtained from data association.Extensive ablation experiments validate the effectiveness of our method and itsgeneralization ability on different detectors. Cross-comparison to otherforecasting schemes further proves the superiority of our method. Code will bereleased upon acceptance.

Trajectory Forecasting from Detection with Uncertainty-Aware Motion Encoding