Multi-scale 3D detection features can be approximated from a single neighborhood using knowledge distillation and class-aware statistics, reducing computation while maintaining accuracy.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Transferable Class Statistics and Multi-scale Feature Approximation for 3D Object Detection
Multi-scale 3D detection features can be approximated from a single neighborhood using knowledge distillation and class-aware statistics, reducing computation while maintaining accuracy.