Ben Saunders

Co-Founder at @SignapseAI, PhD Student at @CogVis, CVSSP, researching Sign Language Production supervised by Richard Bowden and @ncihancamgoz

[Ben Saunders @BenMSaunders]

🥳🥳 Happy to announce our paper "Signing at Scale: Learning to Co-Articulate Signs for Large-Scale Photo-Realistic Sign Language Production" has been accepted at #CVPR2022 #CVPR Work done with @ncihancamgoz and Rich Bowden @CogVis. Paper details to follow.

[#CVPR2026 @CVPR]

We have successfully wrapped up the #CVPR2022 decision process with the tremendous effort of our reviewers and area chairs.

The list of paper IDs of accepted papers are here:
https://t.co/VN6aOCNPeZ

The final reviews and meta reviews will be released in the next day or two. 1/2

[Gül Varol @gulvarol]

One more month to participate in these #ECCV2022 challenges ⏳We are looking forward to numbers better than the baselines!

[Liliane Momeni @LiliMomeni]

We are very excited to announce the SLRTP workshop #ECCV2022 @eccvconf Challenge is open and accepting submissions. Join us to push the boundaries on 3 sign language understanding tasks: sign recognition, subtitle alignment and sign spotting! https://t.co/AO14JaZiOy

[Ben Saunders @BenMSaunders]

@DocXavi @mandacduarte @skamalas @LaiaTarres @JordiTorresAI @ElisaSayrol @asmeaton Congrats!! 🎉

[Mathias Müller @bricksdont]

Major update regarding the WMT shared task on sign language translation: after considerable delay (for which I am very sorry) our training data and baseline code are now available. 🎉 https://t.co/WfLvDUWGWN

[Ben Saunders @BenMSaunders]

My paper "Skeletal Graph Self-Attention: Embedding a Skeleton Inductive Bias into Sign Language Production" was accepted to the "Sign Language Translation and Avatar Technology" workshop at @lrec2022 🥳 Pre-print available at https://t.co/04Y0rwCxPP, find out more at #SLTAT22

[Cognitive Vision Lab @ CVSSP, Surrey @CogVis]

🎉 Pleased to announce we had 2 papers accepted to the #SLTAT22 workshop on "Sign Language Translation and Avatar Technology" at @lrec2022 https://t.co/881lmP7kqc🎉 Congrats to all co-authors: Harry Walsh, @BenMSaunders, @ncihancamgoz and Richard Bowden. Paper details to follow.

[Ben Saunders @BenMSaunders]

The preprint of our CVPR22 paper "Signing at Scale: Learning to Co-Articulate Signs for Large-Scale Photo-Realistic Sign Language Production" is now available on ArXiv!

[DeepAI @DeepAI]

Signing at Scale: Learning to Co-Articulate Signs for Large-Scale Photo-Realistic Sign Language Production
https://t.co/4jUb6T8Fqw
by @BenMSaunders et al. including @ncihancamgoz
#DeepLearning #LossFunction

[Ben Saunders @BenMSaunders]

Opportunities open to join my lab at CVSSP studying a PhD in Sign Language Translation/Production! Let me know if you have any questions about what a PhD here entails😀

[Cognitive Vision Lab @ CVSSP, Surrey @CogVis]

Come join us! 🎉 We have 2 PhD studentships available researching Sign Language Production (https://t.co/nH4IRN9a6D) and Sign Language Translation (https://t.co/pToXlB8dPE) 👋 If interested, send your CV and covering letter to Prof. Richard Bowden at [email protected] 📜����️

[Ben Saunders @BenMSaunders]

@mokhtarmoussa Thanks :) The MSE loss function applied to the skeleton is also applied to the counter, so the decoder learns EOS in parallel to skeleton prediction

[Ben Saunders @BenMSaunders]

Our paper "Progressive Transformers for End-to-End Sign Language Production" is being presented at #ECCV20. Please check out our poster videos and paper at https://t.co/YIv4262fwz

[DeepAI @DeepAI]

Skeletal Graph Self-Attention: Embedding a Skeleton Inductive Bias into Sign Language Production
https://t.co/vRotSa9s2i
by @BenMSaunders et al. including @ncihancamgoz
#NeuralMachineTranslation #Vector

[Ben Saunders @BenMSaunders]

Our paper "Skeletal Graph Self-Attention: Embedding a Skeleton Inductive Bias into Sign Language Production" is available at https://t.co/fTXzaTFHrL. We represent sign language sequences in a skeletal graph structure and propose a novel Skeletal Graph Self-Attention layer for SLP

[Ben Saunders @BenMSaunders]

@IEEEFG2021 More details can be found in our video: https://t.co/elddHsqnJO

[Ben Saunders @BenMSaunders]

I am presenting a poster at the Face & Gesture conference today at 4:30pm GMT. Come along to hear more about our work 😃 @IEEEFG2021 https://t.co/NLzBrlObUA

[Cognitive Vision Lab @ CVSSP, Surrey @CogVis]

@IEEEFG2021 @TaoJiang9 @ncihancamgoz @BenMSaunders @Content4All_prj @IEEEFG2021 Poster at 4pm GMT today: "AnonySign: Novel Human Appearance Synthesis for Sign Language Video Anonymisation" by @BenMSaunders, @ncihancamgoz & Richard Bowden. https://t.co/KuEXs9tNlQ

[Ben Saunders @BenMSaunders]

@Deep__AI @ncihancamgoz Thanks for the plug 😄

[Ben Saunders @BenMSaunders]

If you're at @ICCV_2021, check out our paper "Mixed SIGNals: Sign Language Production via a Mixture of Motion Primitives" - Session 2A & 2B Q&A sessions on Tue 12th 11am GMT and Thur 14th 6pm GMT
Video available at https://t.co/Tk6z2I2FzD

[Ben Saunders @BenMSaunders]

The preprint of our #ICCV2021 paper "Mixed SIGNals: Sign Language Production via a Mixture of Motion Primitives" is now available at https://t.co/ebywJfLYwa

[Ben Saunders @BenMSaunders]

Our paper "Mixed SIGNals: Sign Language Production via a Mixture of Motion Primitives" was accepted at #ICCV21 ☺️☺️ The title puns are getting better!

[ICCV2021 @ICCV_2021]

[DECISIONS] List of accepted papers is now publicly available at:
https://t.co/PHDWoowjC3

1617 papers accepted - 25.9% acceptance rate.
Nothing yet on CMT... hang on for meta-reviews soon.

[DeepAI @DeepAI]

AnonySIGN: Novel Human Appearance Synthesis for Sign Language Video Anonymisation
https://t.co/Es6LB9YOIf
by @BenMSaunders et al. including @ncihancamgoz
#Autoencoder #ComputerVision

[Ben Saunders @BenMSaunders]

@Deep__AI @ncihancamgoz Thanks for the shout-out 🙏 See my profile for more details

[Ben Saunders @BenMSaunders]

@releasethegecko @TheBMVA Are registrations now full? On the website it says "Event currently unavailable. Registration is currently unavailable"

[Ben Saunders @BenMSaunders]

Style consistency over a pose sequence is achieved with our proposed style loss.

[Ben Saunders @BenMSaunders]

My paper "AnonySign: Novel Human Appearance Synthesis for Sign Language Video Anonymisation" has been accepted for publication at Face and Gesture 2021 ☺️☺️ Details to follow!

[Cognitive Vision Lab @ CVSSP, Surrey @CogVis]

Pleased to announce we had 3 papers accepted for publication at Face and Gesture 2021 #FG2021 @IEEEFG2021 Congrats to all authors! 🎉🎉 @ncihancamgoz @BenMSaunders @TaoJiang9

[Ben Saunders @BenMSaunders]

Style code traversal between two original appearances generates novel human appearances using ANONYSIGN.