Automatic Face Understanding: Recognizing Families in Photos
The paper addresses the problem of improving kinship recognition, facial landmark localization, and bias measurement in facial recognition, which are important for computer vision researchers and developers. The contributions are incremental, building upon existing challenges.
This paper introduces the largest database for kinship recognition, FIW, which was labeled using a novel clustering algorithm, and demonstrates significant performance gains over previous datasets. It also proposes an adversarial training framework for facial landmark localization that achieves state-of-the-art results while being robust and compact. Additionally, the paper presents BFW, a dataset to measure bias in facial recognition across ethnicity and gender, revealing non-optimal performance with single thresholds.
We built the largest database for kinship recognition. The data were labeled using a novel clustering algorithm that used label proposals as side information to guide more accurate clusters. Great savings in time and human input was had. Statistically, FIW shows enormous gains over its predecessors. We have several benchmarks in kinship verification, family classification, tri-subject verification, and large-scale search and retrieval. We also trained CNNs on FIW and deployed the model on the renowned KinWild I and II to gain SOTA. Most recently, we further augmented FIW with MM. Now, video dynamics, audio, and text captions can be used in the decision making of kinship recognition systems. We expect FIW will significantly impact research and reality. Additionally, we tackled the classic problem of facial landmark localization. A majority of these networks have objectives based on L1 or L2 norms, which inherit several disadvantages. The locations of landmarks are determined from generated heatmaps from which predicted landmark locations get penalized without accounting for the spread: a high scatter corresponds to low confidence and vice-versa. To address this, we introduced an objective that penalizes for low confidence. Another issue is a dependency on labeled data, which is expensive to collect and susceptible to error. We addressed both issues by proposing an adversarial training framework that leverages unlabeled data to improve model performance. Our method claims SOTA on renowned benchmarks. Furthermore, our model is robust with a reduced size: 1/8 the number of channels is comparable to SOTA in real-time on a CPU. Finally, we built BFW to serve as a proxy to measure bias across ethnicity and gender subgroups, allowing us to characterize FR performances per subgroup. We show performances are non-optimal when a single threshold is used to determine whether sample pairs are genuine.