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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Collaborative perception can substantially boost each agent's perceptionability by facilitating communication among multiple agents. However, temporalasynchrony among agents is inevitable in the real world due to communicationdelays, interruptions, and clock misalignments. This issue causes informationmismatch during multi-agent fusion, seriously shaking the foundation ofcollaboration. To address this issue, we propose CoBEVFlow, anasynchrony-robust collaborative perception system based on bird's eye view(BEV) flow. The key intuition of CoBEVFlow is to compensate motions to alignasynchronous collaboration messages sent by multiple agents. To model themotion in a scene, we propose BEV flow, which is a collection of the motionvector corresponding to each spatial location. Based on BEV flow, asynchronousperceptual features can be reassigned to appropriate positions, mitigating theimpact of asynchrony. CoBEVFlow has two advantages: (i) CoBEVFlow can handleasynchronous collaboration messages sent at irregular, continuous time stampswithout discretization; and (ii) with BEV flow, CoBEVFlow only transports theoriginal perceptual features, instead of generating new perceptual features,avoiding additional noises. To validate CoBEVFlow's efficacy, we createIRregular V2V(IRV2V), the first synthetic collaborative perception dataset withvarious temporal asynchronies that simulate different real-world scenarios.Extensive experiments conducted on both IRV2V and the real-world datasetDAIR-V2X show that CoBEVFlow consistently outperforms other baselines and isrobust in extremely asynchronous settings. The code is available athttps://github.com/MediaBrain-SJTU/CoBEVFlow.

Asynchrony-Robust Collaborative Perception via Bird's Eye View Flow