Ben Saunders
PhD Candidate (University of Surrey)
Centre for Vision,Speech and Signal Processing (CVSSP)
About
I am a 3rd year PhD Student at the Centre for Vision, Speech and Signal Processing (CVSSP) researching Sign Language Production, under the supervision of
Prof. Richard Bowden and Necati Cihan Camgoz. I received my Master's degree with distinction in Artificial Intelligence from Queen Mary University, having achieved a first class honours Bachelor's from London School of Economics.
My PhD has focused on the neural production of sign languages , researching how computer vision and natural language processing can be used to translate spoken language sentences into photo-realistic sign language sequences, aka Sign Language Production.
Feel free to contact me to chat about my research or potential collaborations.
News
- Mar. '22 1 paper accepted to Computer Vision and Pattern Recognition (CVPR '22)!
- Jul. '21 1 paper accepted to International Conference on Computer Vision (ICCV '21)!
- Jul. '21 2 papers accepted to International Conference on Automatic Face and Gesture Recognition 2021 (FG '21, FG '21)
- May. '21 1 paper accepted to an International Journal of Computer Vision (IJCV) Special Issue on Human pose, Motion, Activities and Shape in 3D
- Nov. '20 I successfully defended my 1 year PhD Confirmation
- Sep. '20 1 paper accepted to the British Machine Vision Conference (BMVC '20)
- Aug. '20 1 paper accepted to the European Conference on Computer Vision (ECCV '20)
- Aug. '19 I graduated with a Distinction from my Artificial Intelligence Master's at Queen Mary University
Research
My long-term research goal is to develop cutting-edge technology designed specifically for the disabled, using computer vision and natural language processing. I believe that all people can and should benefit from the current AI technologoical advances.
I am particularly interested in Computational Sign Language research, with a focus on the translation from spoken language sequences to continuous sign language sequences. In addition, I want to move the field of Sign Language Production away from a skeletal representation towards photo-realistic sign language videos.
My current research focus can be summarised as:
- Neural Sign Language Production (ECCV'20, BMVC'20, IJCV'21, ICCV '21, CVPR'22)
- Photo-Realistic Sign Language Video Generation (ArXiv, FG'21, CVPR'22)
Publications
2022
-
Signing at Scale: Learning to Co-Articulate Signs for Large-Scale Photo-Realistic Sign Language Production
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
Computer Vision and Pattern Recognition (CVPR '22)
2021
-
Mixed SIGNals: Sign Language Production via a Mixture of Motion Primitives
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
International Conference on Computer Vision (ICCV '21)
-
ANONYSIGN: Novel Human Appearance Synthesis for Sign Language Video Anonymisation
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
International Conference on Automatic Face and Gesture Recognition 2021 (FG'21)
-
Content4All Open Research Sign Language Translation Datasets
Necati Cihan Camgoz, Ben Saunders, Guillaume Rochette, Marco Giovanelli, Giacomo Inches, Robin Nachtrab-Ribback and Richard Bowden
International Conference on Automatic Face and Gesture Recognition 2021 (FG'21)
-
Continuous 3D Multi-Channel Sign Language Production via Progressive Transformers and Mixture Density Networks
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
JInternational Journal of Computer Vision (IJCV) Special Issue on Human Pose, Motion, Activities and Shape in 3D
-
SeeHear: Signer Diarisation and a new Dataset
Samuel Albanie, Gül Varol, Liliane Momeni, Triantafyllos Afouras, Andrew Brown, Chuhan Zhang, Ernesto Coto, Necati Cihan Camgöz, Ben Saunders, Neil Fox, Abhishek Dutta, Richard Bowden, Bencie Woll and Andrew Zisserman
International Conference on Acoustics, Speech, and Signal Processing (ICASSP '21)
2020
-
Everybody Sign Now: Translating Spoken Language to Photo Realistic Sign Language Video
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
arXiv preprint arXiv:2011.09846
-
Adversarial Training for Multi-Channel Sign Language Production
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
Proceedings of the British Machine Vision Conference (BMVC '20)
-
Progressive Transformers for End-to-End Sign Language Production
Ben Saunders, Necati Cihan Camgoz and Richard Bowden
Proceedings of the European Conference on Computer Vision (ECCV '20)
Code