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

Given the capability of mitigating the long-tail deficiencies andintricate-shaped absence prevalent in 3D object detection, occupancy predictionhas become a pivotal component in autonomous driving systems. However, theprocession of three-dimensional voxel-level representations inevitablyintroduces large overhead in both memory and computation, obstructing thedeployment of to-date occupancy prediction approaches. In contrast to the trendof making the model larger and more complicated, we argue that a desirableframework should be deployment-friendly to diverse chips while maintaining highprecision. To this end, we propose a plug-and-play paradigm, namely FlashOCC,to consolidate rapid and memory-efficient occupancy prediction whilemaintaining high precision. Particularly, our FlashOCC makes two improvementsbased on the contemporary voxel-level occupancy prediction approaches. Firstly,the features are kept in the BEV, enabling the employment of efficient 2Dconvolutional layers for feature extraction. Secondly, a channel-to-heighttransformation is introduced to lift the output logits from the BEV into the 3Dspace. We apply the FlashOCC to diverse occupancy prediction baselines on thechallenging Occ3D-nuScenes benchmarks and conduct extensive experiments tovalidate the effectiveness. The results substantiate the superiority of ourplug-and-play paradigm over previous state-of-the-art methods in terms ofprecision, runtime efficiency, and memory costs, demonstrating its potentialfor deployment. The code will be made available.

FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugin