Additional Baseline Metrics for the paper "Extended YouTube Faces: a Dataset for Heterogeneous Open-Set Face Identification"
This work provides incremental improvements to a dataset for heterogeneous open-set face identification, enhancing reliability for researchers in computer vision.
The authors corrected labeling errors and regenerated the evaluation protocol for the Extended YouTube Faces dataset, then reproduced and extended experimental results with additional standard metrics.
In this report, we provide additional and corrected results for the paper "Extended YouTube Faces: a Dataset for Heterogeneous Open-Set Face Identification". After further investigations, we discovered and corrected wrongly labeled images and incorrect identities. This forced us to re-generate the evaluation protocol for the new data; in doing so, we also reproduced and extended the experimental results with other standard metrics and measures used in the literature. The reader can refer to the original paper for additional details regarding the data collection procedure and recognition pipeline.