Talk:FaceNet
| This is the talk page for discussing improvements to the FaceNet article. This is not a forum for general discussion of the article's subject. |
Article policies
|
| Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
Open Source Re-implementation
[edit] | This edit request by an editor with a conflict of interest was declined. |
Even though the backbone of FaceNet model is shared in the research paper authored by Google, its pre-trained weights is not released publicly. Still, some open source studies re-trained the model with publicly available datasets [1]. That open source re-implementation shows a robust performance when compared to original study, with 98.4% score whereas human-beings have 97.5% score on the same dataset[2]. Johncasey (talk) 11:40, 13 August 2024 (UTC) Johncasey (talk) 11:40, 13 August 2024 (UTC)
- We would need an independent secondary source to write about this. Adding this text would serve only to promote your version based on primary sourcing - we don't do that here. - MrOllie (talk) 11:44, 13 August 2024 (UTC)
References
- ^ Sefik Ilkin Serengil. "DeepFace: A Lightweight Face Recognition and Facial Attribute Analysis (Age, Gender, Emotion and Race) Library for Python". GitHub. Retrieved 2024-08-12.
- ^ Serengil, Sefik; Ozpinar, Alper (2024). "A Benchmark of Facial Recognition Pipelines and Co-Usability Performances of Modules". Journal of Information Technologies. doi:10.17671/gazibtd.1399077.