Use packed_simd to speed up FeatureHeap::add

rust-cv / hwt

Hamming Weight Tree from the paper "Online Nearest Neighbor Search in Hamming Space"

MIT License

7 stars 1 forks source link

Use packed_simd to speed up FeatureHeap::add #7

Closed vadixidav closed 5 years ago

vadixidav commented 5 years ago

Profiling currently shows a lot of time is spent in FeatureHeap::add doing XOR, popcount, and comparations. We can speed this up with SIMD by doing the XOR, popcnt, and compares in parallel.