3,839
edits
Synthetic human-like fakes: Difference between revisions
→2020's synthetic human-like fakes: + The scientists wrote an article on their work titled "Deepfake audio has a tell – researchers use fluid dynamics to spot artificial imposter voices" at theconversation.com that was published Tuesday 2022-09-20
Juho Kunsola (talk | contribs) (→2020's synthetic human-like fakes: + The work "Who Are You (I Really Wanna Know)? Detecting Audio DeepFakes Through Vocal Tract Reconstruction" from researchers of the Florida Institute for Cybersecurity Research (FICS) in the w:University of Florida and received funding from the w:Office of Naval Research and was presented in August 2020 at the w:USENIX Security Symposium.) |
Juho Kunsola (talk | contribs) (→2020's synthetic human-like fakes: + The scientists wrote an article on their work titled "Deepfake audio has a tell – researchers use fluid dynamics to spot artificial imposter voices" at theconversation.com that was published Tuesday 2022-09-20) |
||
| Line 362: | Line 362: | ||
[[File:Appearance of Queen Elizabeth II stolen by Channel 4 in Dec 2020 (screenshot at 191s).png|thumb|right|480px|In Dec 2020 Channel 4 aired a Queen-like fake i.e. they had thieved the appearance of Queen Elizabeth II using deepfake methods.]] | [[File:Appearance of Queen Elizabeth II stolen by Channel 4 in Dec 2020 (screenshot at 191s).png|thumb|right|480px|In Dec 2020 Channel 4 aired a Queen-like fake i.e. they had thieved the appearance of Queen Elizabeth II using deepfake methods.]] | ||
* '''2022''' | '''<font color="green">counter-measure</font>''' | The work [https://www.usenix.org/conference/usenixsecurity22/presentation/blue '''Who Are You (I Really Wanna Know)? Detecting Audio DeepFakes Through Vocal Tract Reconstruction''' at usenix.org with links to the paper, presentation handout and slides] from researchers of the Florida Institute for Cybersecurity Research (FICS) in the [[w:University of Florida]] and received funding from the [[w:Office of Naval Research]] and was presented in August 2020 at the [[w:USENIX]] Security Symposium. | * '''2022''' | '''<font color="green">counter-measure</font>''' | The work [https://www.usenix.org/conference/usenixsecurity22/presentation/blue '''''Who Are You (I Really Wanna Know)? Detecting Audio DeepFakes Through Vocal Tract Reconstruction''''' at usenix.org with links to the paper, presentation handout and slides] from researchers of the Florida Institute for Cybersecurity Research (FICS) in the [[w:University of Florida]] and received funding from the [[w:Office of Naval Research]] and was presented in August 2020 at the [[w:USENIX]] Security Symposium. The scientists wrote an article on their work titled [https://theconversation.com/deepfake-audio-has-a-tell-researchers-use-fluid-dynamics-to-spot-artificial-imposter-voices-189104 '''''Deepfake audio has a tell – researchers use fluid dynamics to spot artificial imposter voices''''' at theconversation.com] that was published Tuesday 2022-09-20. | ||
* '''2022''' | '''<font color="green">counter-measure</font>''' | [https://arxiv.org/pdf/2206.12043.pdf '''Protecting President Zelenskyy against deep fakes''' a 2022 preprint at arxiv.org] by Matyáš Boháček of Johannes Kepler Gymnasium and [[w:Hany Farid]], the dean and head of of [[w:University of California, Berkeley School of Information|w:Berkeley School of Information at the University of California, Berkeley]]. This brief paper describes their automated digital look-alike detection system and evaluate its efficacy and reliability in comparison to humans with untrained eyes. Their work provides automated evaluation tools to catch so called "deep fakes" and their motivation seems to have been to find automation armor against disinformation warfare against humans and the humanity. Automated digital [[Glossary#Media forensics|media forensics]] is a very good idea explored by many. Boháček and Farid 2022 detection system works by evaluating both facial mannerisms as well as gestural mannerisms to detect the non-human ones from the ones that are human in origin. | * '''2022''' | '''<font color="green">counter-measure</font>''' | [https://arxiv.org/pdf/2206.12043.pdf '''''Protecting President Zelenskyy against deep fakes''''' a 2022 preprint at arxiv.org] by Matyáš Boháček of Johannes Kepler Gymnasium and [[w:Hany Farid]], the dean and head of of [[w:University of California, Berkeley School of Information|w:Berkeley School of Information at the University of California, Berkeley]]. This brief paper describes their automated digital look-alike detection system and evaluate its efficacy and reliability in comparison to humans with untrained eyes. Their work provides automated evaluation tools to catch so called "deep fakes" and their motivation seems to have been to find automation armor against disinformation warfare against humans and the humanity. Automated digital [[Glossary#Media forensics|media forensics]] is a very good idea explored by many. Boháček and Farid 2022 detection system works by evaluating both facial mannerisms as well as gestural mannerisms to detect the non-human ones from the ones that are human in origin. | ||
* '''2021''' | Science and demonstration | In the NeurIPS 2021 held virtually in December researchers from Nvidia and [[w:Aalto University]] present their paper [https://nvlabs.github.io/stylegan3/ '''''Alias-Free Generative Adversarial Networks (StyleGAN3)''''' at nvlabs.github.io] and associated [https://github.com/NVlabs/stylegan3 implementation] in [[w:PyTorch]] and the results are deceivingly human-like in appearance. [https://nvlabs-fi-cdn.nvidia.com/stylegan3/stylegan3-paper.pdf StyleGAN3 paper as .pdf at nvlabs-fi-cdn.nvidia.com] | * '''2021''' | Science and demonstration | In the NeurIPS 2021 held virtually in December researchers from Nvidia and [[w:Aalto University]] present their paper [https://nvlabs.github.io/stylegan3/ '''''Alias-Free Generative Adversarial Networks (StyleGAN3)''''' at nvlabs.github.io] and associated [https://github.com/NVlabs/stylegan3 implementation] in [[w:PyTorch]] and the results are deceivingly human-like in appearance. [https://nvlabs-fi-cdn.nvidia.com/stylegan3/stylegan3-paper.pdf StyleGAN3 paper as .pdf at nvlabs-fi-cdn.nvidia.com] | ||