Fatemeh Saleh


Research Scientist
Microsoft, Cambridge, UK


           

profile photo

Bio: I am a Research Scientist at Microsoft in Cambridge, UK, working in Mesh Labs. Before that, I was a research scientist at Samsung AI Center, Cambridge. While I am broadly interested in machine learning and computer vision and their applications in mixed-reality, my main focus is on utilizing machine learning techniques for human behaviour understanding as well as synthetic data generation.


Latest News

One paper accepted in ICLR 2023. Congratulations to Fuwen and the team at Samsung AI Center, Cambridge.
From July 2022, I joined Microsoft Mesh Labs in Cambridge as a Research Scientist.
One paper accepted in CVPR 2022. Congratulations to Mahsa and the team.
One paper on motion prediction accepted at ICCV 2021 (Oral).
From April 2021, I joined Samsung AI Center in Cambridge as a Research Scientist.
One paper accepted in TPAMI. Congradulations to Jing and the team.
One paper accepted in CVPR'21 (oral).
Together with Sadegh Aliakbarian and Stephen Gould, we've been awarded $20,000 grant for AI for Decision Making Initiative.
I have been selected as an outstanding reviewer for ECCV'20.
One paper accepted in ECCV'20. Congradulations to Mahsa.
Our oral CVPR 2020 paper, UC-Net, has been nominated in the finalists of CVPR 2020 paper awards!
Two papers accepted in CVPR'20 (one oral). Congradulations to Sadegh and Jing.
One paper accepted in WACV'20. Congradulations to Cristian Rodriguez.
From January 2019, I joined ANU/ACRV as a Research Fellow under supervision of Prof. Stephen Gould.
After wonderful 3.5 years, I just finished my PhD at ANU! Please find my thesis here.

Publications
Effective Self-supervised Pre-training on Low-compute Networks without Distillation
Fuwen Tan, Fatemeh Saleh, Brais Martinez
International Conference on Learning Representations (ICLR), 2023
[project page / arXiv / video ]
JRDB-Act: A Large-scale Dataset for Spatio-temporal Action, Social Group and Activity Detection
Mahsa Ehsanpour, Fatemeh Saleh, Silvio Savarese, Ian Reid, Hamid Rezatofighi
International Conference on Learning Representations (ICLR), 2023
[project page / arXiv / video ]
Contextually Plausible and Diverse 3D Human Motion Prediction
Sadegh Aliakbarian, Fatemeh Saleh, Lars Petersson, Mathieu Salzmann, Stephen Gould
International Conference on Computer Vision (ICCV), 2021 (Oral presentation)
[project page / arXiv / video ]
Uncertainty Inspired RGB-D Saliency Detection
Jing Zhang, Deng-Ping Fan, Yuchao Dai, Saeed Anwar, Fatemeh Saleh, Sadegh Aliakbarian, Nick Barnes
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021
[project page / arXiv / video / code]
Probabilistic Tracklet Scoring and Inpainting for Multiple Object Tracking
Fatemeh Saleh, Sadegh Aliakbarian, Hamid Rezatofighi, Mathieu Salzmann, Stephen Gould
Proceedings of the IEEE international conference on computer vision (CVPR), 2021 (Oral presentation)
[project page / arXiv / video / code]
The IKEA ASM Dataset: Understanding People Assembling Furniture through Actions, Objects and Pose
Yizhak Ben-Shabat, Xin Yu, Fatemeh Saleh, Dylan Campbell, Cristian Rodriguez-Opazo, Hongdong Li, Stephen Gould
Winter Conference on Applications of Computer Vision (WACV), 2021
[project page / arXiv / Dataset / code]
Joint learning of Social Groups, Individuals Action and Sub-group Activities in Videos
Mahsa Ehsanpour, Alireza Abedin, Fatemeh Saleh, Javen Shi, Ian Reid, Hamid Rezatofighi
European Conference on Computer Vision (ECCV), 2020
[project page / arXiv / video / code]
A Stochastic Conditioning Scheme for Diverse Human Motion Prediction
Sadegh Aliakbarian, Fatemeh Saleh, Mathieu Salzmann, Lars Petersson, Stephen Gould
Proceedings of the IEEE international conference on computer vision (CVPR), 2020
[project page / arXiv / video / code]
UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders
Jing Zhang, Deng-Ping Fan, Yuchao Dai, Saeed Anwar, Fatemeh Saleh, Tong Zhang, Nick Barnes
Proceedings of the IEEE international conference on computer vision (CVPR), 2020 (Oral presentation)
[project page / arXiv / video / code]
Proposal-free Temporal Moment Localization of a Natural-Language Query in Video using Guided Attention
Cristian Rodriguez Opazo, Edison Marrese-Taylor, Fatemeh Saleh, Hongdong Li, Stephen Gould
Winter Conference on Applications of Computer Vision (WACV), 2020
[project page / arXiv / Video / Code]
Sampling Good Latent Variables via CPP-VAEs: VAEs with Condition Posterior as Prior
Sadegh Aliakbarian, Fatemeh Saleh, Mathieu Salzmann, Lars Petersson, Stephen Gould
ArXiv, 2019
[project page / arXiv / Video / Code]
VIENA2: A Driving Anticipation Dataset
Sadegh Aliakbarian, Fatemeh Saleh, Mathieu Salzmann, Basura Fernando, Lars Petersson, Lars Andersson
Asian Conference on Computer Vision (ACCV), 2018
[project page / arXiv / Video / Code]
Effective Use of Synthetic Data in Urban Scene Semantic Segmentation
Fatemeh Saleh, Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Jose M Alvarez
European Conference on Computer Vision (ECCV), 2018
[project page / arXiv / Video / Code]
Incorporating Network Built-in Priors in Weakly-supervised Semantic Segmentation
Fatemeh Saleh, Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Jose M Alvarez, Stephen Gould
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2018
[project page / arXiv / Video / Code]
Encouraging LSTMs to Anticipate Actions Very Early
Sadegh Aliakbarian, Fatemeh Saleh, Mathieu Salzmann, Basura Fernando, Lars Petersson, Lars Andersson
International Conference on Computer Vision (ICCV), 2017
[project page / arXiv / Video / Code]
Bringing background into the foreground: Making all classes equal in weakly-supervised video semantic segmentation
Fatemeh Saleh, Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Jose M Alvarez
International Conference on Computer Vision (ICCV), 2017
[project page / arXiv / Video / Code]
Built-in Foreground/Background Prior for Weakly-Supervised Semantic Segmentation
Fatemeh Saleh, Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Stephen Gould, Jose M Alvarez
European Conference on Computer Vision (ECCV), 2016
[project page / arXiv / Video / Code]
Recent Experiences
Research Scientist | Microsoft, Cambridge, UK
[July 2022 - Now]
Working on Microsoft Mesh and Mesh for Teams.
Research Scientist | Samsung AI Center, Cambridge, UK
[April 2021 - July 2022]
Working on self-supervised representation learning.
Research Fellow | Australian Centre for Robotic Vision (ACRV), Canberra, Australia
[January 2019 - March 2021]
Working on analysing human motion and behaviour, e.g., multiple human tracking, diverse motion prediction, social activity analysis, and action forecasting.
Machine Learning Research Consultant | Joint ACRV/LumaChain collaboration
[June 2020 - August 2020]
As a research consultant, we are delivering multiple (deformable and visually similar) object tracking in highly occluded scenarios. The project is related to the real-world challenging scenarios of food chain industry.
Research Intern | Qualcomm AI Research, Amsterdam, Netherlands
[May 2018 - October 2018]
Working on self-supervised representation learning for video analysis tasks.
Education
PhD, Computer Science, Australian National University, [June 2015 - February 2019]
MSc, Artificial Intelligence, Sharif University of Technology, [October 2011 - September 2013]
BSc, Computer Software Engineering, Isfahan University of Technology, [October 2007 - September 2011]

 

Thanks for the template