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

To ensure the safe and efficient navigation of autonomous vehicles andadvanced driving assistance systems in complex traffic scenarios, predictingthe future bounding boxes of surrounding traffic agents is crucial. However,simultaneously predicting the future location and scale of target trafficagents from the egocentric view poses challenges due to the vehicle's egomotioncausing considerable field-of-view changes. Moreover, in anomalous or riskysituations, tracking loss or abrupt motion changes limit the availableobservation time, requiring learning of cues within a short time window.Existing methods typically use a simple concatenation operation to combinedifferent cues, overlooking their dynamics over time. To address this, thispaper introduces the Fusion-Gated Recurrent Unit (Fusion-GRU) network, a novelencoder-decoder architecture for future bounding box localization. Unliketraditional GRUs, Fusion-GRU accounts for mutual and complex interactions amonginput features. Moreover, an intermediary estimator coupled with aself-attention aggregation layer is also introduced to learn sequentialdependencies for long range prediction. Finally, a GRU decoder is employed topredict the future bounding boxes. The proposed method is evaluated on twopublicly available datasets, ROL and HEV-I. The experimental results showcasethe promising performance of the Fusion-GRU, demonstrating its effectiveness inpredicting future bounding boxes of traffic agents.

Fusion-GRU: A Deep Learning Model for Future Bounding Box Prediction of Traffic Agents in Risky Driving Videos