google/quickshift
A clustering algorithm that first finds the high-density regions (cluster-cores) of the data and then clusters the remaining points by hill-climbing. Such seedings act as more stable and expressive cluster-cores than the singleton modes found by popular algorithm such as mean shift. (https://arxiv.org/abs/1805.07909)
C++Apache-2.0
Stargazers
- adaburrowsUSA
- azazdeazTokyo, Japan
- bill-kalog
- codealphago
- dbremnerSeattle
- dimitryslavinSan Francisco, CA
- dreamplayerzhang
- dsx-aishanghai
- ducminhkhoiOregon State University
- eddingsArizona
- fly51flyPRIS
- gracelinmumu银河系
- HLT
- horsto@kavli-ntnu
- kedartalK2DQ
- kimnwang
- KotsakisStockholm, Sweden
- KvaroverHeavenlocalhost, Taiwan
- laudLau Brothers LLC
- marcusrbrown
- minghao2016
- pbamotraAutodesk, Ex-Coupang, CMU-Pittsburgh
- SanghyukChun@naver-ai lab
- sarnthilalcemy
- sguzmanBakersfield, CA
- sikofittStanford University, Electrical Engineering Department
- sirgatezByteDance
- sivapvarmaUC San Diego
- StanpolOslo
- stormraiserUSC
- theRealSuperMariohttps://modugen.de
- tnakaeBizReach, Inc.
- unyqhz
- vdaghanHacettepe University
- wangfan233Airbnb
- xuanhan863Los Angeles, USA