In Real-World Computer Vision from Inputs with Limited Quality Workshop (RLQ) in conjunction with ICCV 2021
We investigate ways to leverage uncertainty in face images to improve the quality of the face clusters. We observe that popular clustering algorithms do not produce better quality clusters when clustering probabilistic face represen- tations that implicitly model uncertainty – these algorithms predict up to 9.6X more clusters than the ground truth for the IJB-A benchmark. We empirically analyze the causes for this unexpected behavior and identify excessive false- positives and false-negatives (when comparing face-pairs) as the main reasons for poor quality clustering. Based on this insight, we propose an uncertainty-aware clustering al- gorithm, UAC, which explicitly leverages uncertainty infor- mation during clustering to decide when a pair of faces are similar or when a predicted cluster should be discarded. UAC considers (a) uncertainty of faces in face-pairs, (b) bins face-pairs into different categories based on an un- certainty threshold, (c) intelligently varies the similarity threshold during clustering to reduce false-negatives and false-positives, and (d) discards predicted clusters that ex- hibit a high measure of uncertainty. Extensive experimen- tal results on several popular benchmarks and comparisons with state-of-the-art clustering methods show that UAC pro- duces significantly better clusters by leveraging uncertainty in face images – predicted number of clusters is up to 0.18X more of the ground truth for the IJB-A benchmark.
divided by Adrien Coquet from the Noun Project