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

We present \textit{VoxelKP}, a novel fully sparse network architecturetailored for human keypoint estimation in LiDAR data. The key challenge is thatobjects are distributed sparsely in 3D space, while human keypoint detectionrequires detailed local information wherever humans are present. We proposefour novel ideas in this paper. First, we propose sparse selective kernels tocapture multi-scale context. Second, we introduce sparse box-attention to focuson learning spatial correlations between keypoints within each human instance.Third, we incorporate a spatial encoding to leverage absolute 3D coordinateswhen projecting 3D voxels to a 2D grid encoding a bird's eye view. Finally, wepropose hybrid feature learning to combine the processing of per-voxel featureswith sparse convolution. We evaluate our method on the Waymo dataset andachieve an improvement of $27\%$ on the MPJPE metric compared to thestate-of-the-art, \textit{HUM3DIL}, trained on the same data, and $12\%$against the state-of-the-art, \textit{GC-KPL}, pretrained on a $25\times$larger dataset. To the best of our knowledge, \textit{VoxelKP} is the firstsingle-staged, fully sparse network that is specifically designed foraddressing the challenging task of 3D keypoint estimation from LiDAR data,achieving state-of-the-art performances. Our code is available at\url{https://github.com/shijianjian/VoxelKP}.

VoxelKP: A Voxel-based Network Architecture for Human Keypoint Estimation in LiDAR Data