AI Paper Atlas

Research map / Learning theory and algorithms / Clustering

Clustering Millions of Faces by Identity

Charles Otto, Dayong Wang, Anil K. Jain

TPAMI 2017 · Learning theory and algorithms › Clustering · approach: Clustering approximation algorithms · 165 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use approximate rank-order clustering instead of original rank-orderface clustering · LFW (full)F-measure: 0.80 -> 0.87Improved
use approximate rank-order instead of k-meansface clustering · LFW (full)F-measure: 0.36 -> 0.87Improved
use randomized k-d tree approximate neighbors instead of brute-force (ablation)approx. rank-order · LFWF-measure: 0.72 -> 0.87Improved
use randomized k-d tree approximate neighbors instead of brute-force (ablation)approx. rank-order · LFW + 1MF-measure: 0.49 -> 0.79Improved
use randomized k-d tree instead of Chen et al. kNN graph (ablation)approx. rank-order · LFW + 1MF-measure: 0.41 -> 0.79Improved
use randomized k-d tree instead of brute-force neighbors (ablation)approx. rank-order · LFW + 1Mrun-time: 14:18:24 -> 00:07:20Improved
increase k-d tree search size linearly instead of fixed 2,000 (ablation)approx. rank-order · LFW + 5MF-measure: 0.13 -> 0.67Improved
increase search size logarithmically instead of linearly (ablation)approx. rank-order · LFW + 5MF-measure: 0.67 -> 0.33Hurt
add more distractor images (1M to 123M)approx. rank-order · LFW + distractorsF-measure: 0.87 -> 0.27Hurt
exclude clusters of size 2 from cluster ranking (ablation)per-cluster quality measure · LFWcorrelation with precision: 0.427 -> 0.460Improved
combine intra and inter-cluster edge distance instead of inter alone (ablation)per-cluster quality measure · LFWcorrelation with precision: 0.325 -> 0.427Improved
use graph-based measures (coverage, MQ) instead of distance-based measures (ablation)per-cluster quality measure · LFWcorrelation with precision: 0.117 -> 0.325Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy