Person Recognition in Personal Photo Collections

By Seong Joon Oh, Rodrigo Benenson, Mario Fritz, Bernt Schiele. 2015

55 citations recorded by the source corpus.

Recognising persons in everyday photos presents major challenges (occluded faces, different clothing, locations, etc.) for machine vision. We propose a convnet based person recognition system on which we provide an in-depth analysis of informativeness of different body cues, impact of training data, and the common failure modes of the system. In addition, we discuss the limitations of existing benchmarks and propose more challenging ones. Our method is simple and is built on open source and open data, yet it improves the state of the art results on a large dataset of social media photos (PIPA).

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

A Hybrid Model for Identity Obfuscation by Face Replacement (2018, methodology): “ results before and after inpainting. On the top row, rendered faces are overlayed onto the color images for better comparison of details. 4.1 Machine recognizers We use an automatic recognizer naeil [39], the state-of-the-art for person recognition in social media images [1][40]. In contrast to typ”
Faceless Person Recognition: Privacy Implications in Social Media (2016, methodology): “Pose-independent descriptors have been explored for the body region [28, 38, 39, 29, 30].”
I-Pic: A Platform for Privacy-Compliant Image Capture (2016, methodology): “While our prior work [7] uses five different body regions (face, head, upper/full body, and scene) to maximize the performance, we only extract features from the face region, and denote this cue as FNet.”
A Domain Based Approach to Social Relation Recognition (2017, methodology): “To model semantic head/body attributes, we choose two image regions based on the groundtruth annotation of PIPA: the head region corresponding to the groundtruth annotation and the body region, defined as 3×head width, 6×head height, following [44].”
Multi-cue Zero-Shot Learning with Strong Supervision (2016, background): “Such information provides strong visual supervision and has shown to greatly improve recognition accuracy [51, 35].”
Adversarial Image Perturbation for Privacy Protection A Game Theory Perspective (2017, methodology): “The person recognition model f is built on a state of the art framework [27].”
Natural and Effective Obfuscation by Head Inpainting (2017, methodology): “Following [16], we ﬁrst train feature extractors over head and body regions, and then train an SVM identity classi-ﬁer on top of those features.”
Person Recognition in Personal Photo Collections (2017, methodology): “[2], the conference version of this paper, have proposed amuch simplermodel naeil that extracts AlexNet cues from multiple fixed image regions.”
When Machine Learning Meets Privacy (2020, background): “ck tool here. •The re-identification attack can be launched by face recognition techniques. The recent advanceinDNNmakesitmoreharmfulfromtwoaspects.First,theprocessbecomesautomatic with high accuracy [67, 151, 168]. Second, traditional protection schemes such as obfuscation no longer work effectivel”