BiPointNet: Binary Neural Network for Point Clouds
Beihang University
|
SenseTime Research
|
UC San Diego
|
|
To alleviate the resource constraint for real-time point cloud applications that
run on edge devices, in this paper we present BiPointNet, the first model binarization
approach for efficient deep learning on point clouds. We first discover that the
immense performance drop of binarized models for point clouds mainly stems from
two challenges: aggregation-induced feature homogenization that leads to a
degradation of information entropy, and scale distortion that hinders optimization
and invalidates scale-sensitive structures. With theoretical justifications and
in-depth analysis, our BiPointNet introduces Entropy-Maximizing Aggregation (EMA)
to modulate the distribution before aggregation for the maximum information entropy,
and Layer-wise Scale Recovery (LSR) to efficiently restore feature representation capacity.
Extensive experiments show that BiPointNet outperforms existing binarization methods by
convincing margins, at the level even comparable with the full precision counterpart.
We highlight that our techniques are generic, guaranteeing significant improvements on
various fundamental tasks and mainstream backbones, e.g., BiPointNet gives an
impressive 14.7× speedup and 18.9× storage saving on real-world resource-constrained devices.
Baselines
Deployment Efficiency on ARM Devices
Acknowledgements
This work was supported by National Natural Science Foundation of China (62022009, 61872021),
Beijing Nova Program of Science and Technology (Z191100001119050), and State Key Lab of Software
Development Environment (SKLSDE-2020ZX-06).
|
Total Visitor Count:
     
Unique Visitor Count:
| |