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/

Zendy Plus

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

Zendy Tools

Zendy Open

Predicting the future occupancy states of the surrounding environment is avital task for autonomous driving. However, current best-performingsingle-modality methods or multi-modality fusion perception methods are onlyable to predict uniform snapshots of future occupancy states and requirestrictly synchronized sensory data for sensor fusion. We propose a novelframework, StreamingFlow, to lift these strong limitations. StreamingFlow is anovel BEV occupancy predictor that ingests asynchronous multi-sensor datastreams for fusion and performs streaming forecasting of the future occupancymap at any future timestamps. By integrating neural ordinary differentialequations (N-ODE) into recurrent neural networks, StreamingFlow learnsderivatives of BEV features over temporal horizons, updates the implicitsensor's BEV features as part of the fusion process, and propagates BEV statesto the desired future time point. It shows good zero-shot generalizationability of prediction, reflected in the interpolation of the observedprediction time horizon and the reasonable inference of the unseen fartherfuture period. Extensive experiments on two large-scale datasets, nuScenes andLyft L5, demonstrate that StreamingFlow significantly outperforms previousvision-based, LiDAR-based methods, and shows superior performance compared tostate-of-the-art fusion-based methods.

StreamingFlow: Streaming Occupancy Forecasting with Asynchronous Multi-modal Data Streams via Neural Ordinary Differential Equation