· 2014
DeepFace: Closing the Gap to Human-Level Performance in Face Verification
- 6,629citations
- 2014year
Short summary
DeepFace, a nine-layer deep neural network with over 120 million parameters, achieves 97.35% accuracy on the Labeled Faces in the Wild (LFW) dataset, surpassing the previous state-of-the-art by over 27% and nearing human-level performance.
AI-generated from the title and abstract; the full text is not read.
Key points
- DeepFace uses a nine-layer deep neural network with over 120 million parameters, employing locally connected layers without weight sharing.
- The network was trained on the largest facial dataset to date, comprising 4 million images from over 4,000 identities.
- A piecewise affine transformation based on explicit 3D face modeling was used for alignment.
- The method achieved 97.35% accuracy on the Labeled Faces in the Wild (LFW) dataset, a 27% reduction in error compared to prior state-of-the-art.
AI-generated from the title and abstract; the full text is not read.
Abstract
In modern face recognition, the conventional pipeline consists of four stages: detect => align => represent => classify. We revisit both the alignment step and the representation step by employing explicit 3D face modeling in order to apply a piecewise affine transformation, and derive a face representation from a nine-layer deep neural network. This deep network involves more than 120 million parameters using several locally connected layers without weight sharing, rather than the standard convolutional layers. Thus we trained it on the largest facial dataset to-date, an identity labeled dataset of four million facial images belonging to more than 4, 000 identities. The learned representations coupling the accurate model-based alignment with the large facial database generalize remarkably well to faces in unconstrained environments, even with a simple classifier. Our method reaches an accuracy of 97.35% on the Labeled Faces in the Wild (LFW) dataset, reducing the error of the current state of the art by more than 27%, closely approaching human-level performance.
The authors' abstract, as published at the source., 2014 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionComputer Science