The MegaFace Benchmark: 1 Million Faces for Recognition at Scale
This work addresses the need for large-scale evaluation in face recognition for researchers and developers, though it is incremental as it builds on existing benchmarks by scaling up data.
The authors tackled the problem of evaluating face recognition algorithms at a million-scale by introducing the MegaFace dataset and challenge, revealing significant performance differences among algorithms that performed similarly on smaller benchmarks.
Recent face recognition experiments on a major benchmark LFW show stunning performance--a number of algorithms achieve near to perfect score, surpassing human recognition rates. In this paper, we advocate evaluations at the million scale (LFW includes only 13K photos of 5K people). To this end, we have assembled the MegaFace dataset and created the first MegaFace challenge. Our dataset includes One Million photos that capture more than 690K different individuals. The challenge evaluates performance of algorithms with increasing numbers of distractors (going from 10 to 1M) in the gallery set. We present both identification and verification performance, evaluate performance with respect to pose and a person's age, and compare as a function of training data size (number of photos and people). We report results of state of the art and baseline algorithms. Our key observations are that testing at the million scale reveals big performance differences (of algorithms that perform similarly well on smaller scale) and that age invariant recognition as well as pose are still challenging for most. The MegaFace dataset, baseline code, and evaluation scripts, are all publicly released for further experimentations at: megaface.cs.washington.edu.