[visionlist] Post-doctorate Offer : Face Inference from Voice

[lezoray]

Post-doctorate Position : Face inference from voice

Context
--------------
The laboratory GREYC UMR CNRS of the University of Caen Normandy and ENSICAEN (Caen, France), in colla- boration with the company United Biometrics (Caen, France), is launching a call for applications for a postdoctoral research position (duration of 3 years) on face inference from voice. The work will be done within the framework of the BIOPOP (BIOmétrie Pour les Opérations) project funded by the AID (Agence Innovation Défense).


Missions
--------------
The goal is to infer information about a person’s face from a raw recording of his or her voice. Recent preliminary works [1, 2, 3, 4] have shown the feasibility of this inference. The aim is not to generate the exact face corresponding to the voice, but to generate a face that allows to highlight the main discriminating characteristics of the face (gender, age, ethnicity, craniofacial attributes). This can be of great interest for security applications. Indeed, inferring a face from a voice can then allow an operator to perform different tasks from the inferred face image. We can mention for example : the verification of the coherence between a voice and a face, the search of the inferred face in a database.
The post-doctoral fellow will carry out a precise state of the art of face inference methods from voice. He will implement a state of the art solution based on generative models. Finally, he will develop a new and more efficient generative model that can guarantee the generation of a realistic face to meet the expectations of the BIOPOP project.

Skills
--------------
—	Ph.D. in computer science and specialized in machine learning.
—	Solid knowledge of deep learning, computer vision.
—	Publications in major conferences in the field.
—	Strong software development/programming skills, especially in Python/PyTorch.
—	Good written and verbal communication skills are required, the candidate must be fluent in French and proficient in written English.
—	Interpersonal skills and the ability to work individually or as a member of a project team are recommended.


General Information
--------------
—	Research Laboratory : The laboratory GREYC (UMR CNRS 6072) is a Joint Research Unit in digital sciences under the supervision of ENSICAEN, CNRS and the University of Caen Normandy (UNICAEN). The work will be carried out within the Image team whose research activities are focused on the development of new methods of processing and analysis of signals/images/videos.
—	Place : Caen (France), located in the Normandy region, near the sea and about 240 km west of Paris, the city still has many old neighborhoods, a population of about 120,000 and an agglomeration of about 250,000 inhabitants, including more than 30,000 students at the University.
—	Duration : 36 months.
—	Salary : About 2900€ gross per month.
—	To apply : Interested candidates should submit (by email, in a single pdf file) their curriculum vitae, list of publications, a cover letter, and contact information for three references (do not include letters of refe- rence with your applications as we will only request them from short-listed candidates). Applications will be accepted until the position is filled. The position will begin in early October.
—	Contact / supervision :
	—	Olivier Lézoray (olivier.lezoray at unicaen.fr, Full Professor, UNICAEN, GREYC)
	—	Sébastien Bougleux (sebastien.bougleux at unicaen.fr, Associate Professor, UNICAEN, GREYC)
	—	Christophe Charrier (christophe.charrier at unicaen.fr, Associate Professor (Habilited), UNICAEN, GREYC)


References
--------------
[1]	Amanda Cardoso Duarte, Francisco Roldan, Miquel Tubau, Janna Escur, Santiago Pascual, Amaia Salvador, Eva Mohedano, Kevin McGuinness, Jordi Torres, and Xavier Giró-i-Nieto, “Wav2pix : Speech-conditioned face generation using generative adversarial networks,” in IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2019, Brighton, United Kingdom, May 12-17, 2019. 2019, pp. 8633–8637, IEEE.
[2]	Zheng Fang, Zhen Liu, Tingting Liu, Chih-Chieh Hung, Jiangjian Xiao, and Guangjin Feng, “Facial expression GAN for voice- driven face generation,” Vis. Comput., vol. 38, no. 3, pp. 1151–1164, 2022.
[3]	Tae-Hyun Oh, Tali Dekel, Changil Kim, Inbar Mosseri, William T. Freeman, Michael Rubinstein, and Wojciech Matusik, “Speech2face : Learning the face behind a voice,” in IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2019, Long Beach, CA, USA, June 16-20, 2019. 2019, pp. 7539–7548, Computer Vision Foundation / IEEE.
[4]	Yandong Wen, Bhiksha Raj, and Rita Singh, “Face reconstruction from voice using generative adversarial networks,” in Advances in Neural Information Processing Systems 32 : Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, December 8-14, 2019, Vancouver, BC, Canada, Hanna M. Wallach, Hugo Larochelle, Alina Beygelzimer, Florence d’Alché-Buc, Emily B. Fox, and Roman Garnett, Eds., 2019, pp. 5266–5275.

Olivier LÉZORAY
Full Professor of Computer Science 

University of Caen Normandy
West Normandy Institute of Technology
Multimedia and Internet Department
F-50000 SAINT-LÔ+33(0)233775514 <tel:+33 2 33 77 55 14>	GREYC UMR CNRS 6072
Image Team - ENSICAEN
6 Bd. Marechal Juin
F-14000 CAEN+33(0)231452927 <tel:+33 2 31 45 29 27>
 <https://linkedin.com/in/olivier-lezoray-0983114/>	 <skype:olezoray>https://lezoray.users.greyc.fr <https://lezoray.users.greyc.fr/>
 <https://unicaen.fr/>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20220525/a897dc5d/attachment.html>