A point transformer that combines perfect spatial hashing with FlashAttention to align geometric neighborhoods with GPU memory tiles achieves 2.25x faster inference and better semantic segmentation than PTv3.
4d spatio-temporal convnets: Minkowski convolutional neural networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Flash3D: Super-scaling Point Transformers through Joint Hardware-Geometry Locality
A point transformer that combines perfect spatial hashing with FlashAttention to align geometric neighborhoods with GPU memory tiles achieves 2.25x faster inference and better semantic segmentation than PTv3.