Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2312.13240 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913362056577024 |
|---|---|
| author | Rozner, Amit Battash, Barak Lindenbaum, Ofir Wolf, Lior |
| author_facet | Rozner, Amit Battash, Barak Lindenbaum, Ofir Wolf, Lior |
| contents | We study the problem of performing face verification with an efficient neural model $f$. The efficiency of $f$ stems from simplifying the face verification problem from an embedding nearest neighbor search into a binary problem; each user has its own neural network $f$. To allow information sharing between different individuals in the training set, we do not train $f$ directly but instead generate the model weights using a hypernetwork $h$. This leads to the generation of a compact personalized model for face identification that can be deployed on edge devices. Key to the method's success is a novel way of generating hard negatives and carefully scheduling the training objectives. Our model leads to a substantially small $f$ requiring only 23k parameters and 5M floating point operations (FLOPS). We use six face verification datasets to demonstrate that our method is on par or better than state-of-the-art models, with a significantly reduced number of parameters and computational burden. Furthermore, we perform an extensive ablation study to demonstrate the importance of each element in our method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_13240 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Efficient Verification-Based Face Identification Rozner, Amit Battash, Barak Lindenbaum, Ofir Wolf, Lior Computer Vision and Pattern Recognition I.4 We study the problem of performing face verification with an efficient neural model $f$. The efficiency of $f$ stems from simplifying the face verification problem from an embedding nearest neighbor search into a binary problem; each user has its own neural network $f$. To allow information sharing between different individuals in the training set, we do not train $f$ directly but instead generate the model weights using a hypernetwork $h$. This leads to the generation of a compact personalized model for face identification that can be deployed on edge devices. Key to the method's success is a novel way of generating hard negatives and carefully scheduling the training objectives. Our model leads to a substantially small $f$ requiring only 23k parameters and 5M floating point operations (FLOPS). We use six face verification datasets to demonstrate that our method is on par or better than state-of-the-art models, with a significantly reduced number of parameters and computational burden. Furthermore, we perform an extensive ablation study to demonstrate the importance of each element in our method. |
| title | Efficient Verification-Based Face Identification |
| topic | Computer Vision and Pattern Recognition I.4 |
| url | https://arxiv.org/abs/2312.13240 |