Sadegh Aliakbarian’s homepage

sadegh[dot]aliakbarian[at]anu[dot]edu[dot]au
Level G, Building 801, CSIRO, Black Mountain Lab, Canberra, Australia
College of Engineering and Computer Science, Australian National University, Canberra, Australia

You can also find me on

Short bio

I'm a last year PhD student at the Australian National University and Australian Centre for Robotic Vision (ACRV). I'm honored to work under supervision of Dr. Lars Petersson (ANU/CSIRO), Dr. Mathieu Salzmann (EPFL), and Dr. Basura Fernando (ACRV/A-STAR) and have pieces of advice from Dr. Stephen Gould (ANU/ACRV). I am now with Five AI (Oxford Research Group) as a research intern, working on adversarial example and model robustness to such examples. I also spent some time at Qualcomm AI Research in Amsterdam as a deep learning research engineering intern, focusing on deep generative models to learn stochasticity in human motion. My areas of interests are in Machine Learning and AI, with focus on generative models and deep learning approaches for computer vision problems. Find out more about my research here.

News

  • One paper in CVPR 2020!
  • I'm joining Five AI (Oxford Research Group) as a research intern, working on adversarial examples.
  • One paper in WACV 2020! Congrats to Mehrdad.
  • One paper accepted at ACCV 2018! VIENA2 dataset is now publicly available.
  • [April 2018] I'm now a deep learning research/engineering intern at Qualcomm AI Research, Amsterdam, Netherlands
  • [July 2017] I'm now an associated AI/ML researcher at Australian Centre for Robotics Vision (ACRV), Canberra, Australia
  • One paper accepted in ECCV 2018 (with the VEIS dataset)!
  • One paper in TPAMI 2018!
  • Two papers in ICCV 2017!
  • One paper in ECCV 2016!

Blog posts


Recent experiences

  • Research Intern | FiveAI | Oxford, UK (January 2020 - ~July 2020)
  • Research Intern | Qualcomm AI Research | Amsterdam, Netherlands (May 2018 - October 2018)
  • Associate Researcher | Australian Centre for Robotic Vision (ACRV), Canberra, Australia (November 2017 - Now)
  • Research Assistant | Smart Vision System, Data61, CSIRO, Canberra, Australia (July 2016 - Now)
  • Research Intern | National ICT Australia (NICTA), Canberra, Australia (June 2015 - March 2016)

Eductation

  • PhD in Computer Science | The Australian National University, Canberra, Australia (July 2016 - Now)
  • B.Sc. in Computer Software Engineering | Isfahan University of Technology, Isfahan, Iran (October 2009 - September 2013)

Recent publications (2016-Now)

A Stochastic Conditioning Scheme for Diverse Human Motion Prediction
Sadegh Aliakbarian, Fatemeh Saleh, Mathieu Salzmann, Lars Petersson, Stephen Gould, Amir Habibian
CVPR 2020
PDF Bibtex Abstract
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
PDF Bibtex Abstract
Super-resolved Chromatic Mapping of Snapshot Mosaic Image Sensors via a Texture Sensitive Residual Network
Mehrdad Shoeiby, Lars Petersson, Mohammad Ali Armin, Sadegh Aliakbarian, Antonio Robles-Kelly
Winter Conference on Applications of Computer Vision (WACV 2020)
PDF Bibtex Abstract
Learning Variations in Human Motion via Mix-and-Match Perturbation
Sadegh Aliakbarian, Fatemeh Saleh, Mathieu Salzmann, Lars Petersson, Stephen Gould, Amir Habibian
ArXiv, 2019
Video results of Mix-and-Match approach is available here.
PDF Bibtex Abstract
VIENA2: A Driving Anticipation Dataset
Mohammad Sadegh Aliakbarian, Fatemehsadat Saleh, Mathieu Salzmann, Basura Fernando, Lars Petersson, Lars Andersson
Asian Conference on Computer Vision (ACCV 2018)
The VIENA2 dataset is available here.
PDF Bibtex Abstract
Effective Use of Synthetic Data in Urban Scene Semantic Segmentation
Fatemeh Sadat Saleh, Mohammad Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Jose M. Alvarez
European Conference on Computer Vision (ECCV 2018)
The VEIS dataset is available here.
PDF Bibtex Abstract
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)
PDF Bibtex Abstract
Encouraging LSTMs to Anticipate Actions Very Early
Mohammad Sadegh Aliakbarian, Fatemehsadat Saleh, Mathieu Salzmann, Basura Fernando, Lars Petersson and Lars Andersson
International Conference on Computer Vision (ICCV 2017)
State-of-the-art in action anticipation!
PDF Bibtex Abstract
Deep Action- and Context-Aware Sequence Learning for Activity Recognition and Anticipation
Mohammad Sadegh Aliakbarian, Fatemehsadat Saleh, Mathieu Salzmann, Basura Fernando, Lars Petersson and Lars Andersson
ArXiv 2016
PDF Bibtex Abstract
Bringing background into the foreground: Making all classes equal in weakly-supervised video semantic segmentation
Fatemeh Sadat Saleh, Mohammad Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Jose M Alvarez
International Conference on Computer Vision (ICCV 2017)
The first weakly-supervised video semantic segmentation approach.
PDF Bibtex Abstract
Built-in Foreground/Background Prior for Weakly-Supervised Semantic Segmentation
Fatemehsadat Saleh, Mohammad Sadegh Aliakbarian, Mathieu Salzmann, Lars Petersson, Stephen Gould, Jose M. Alvarez
European Conference on Computer Vision (ECCV 2016)
PDF Bibtex Abstract

Scholarships and Awards

  • Recipient of €18K grant for R&D from Qualcomm AI Research, 2018
  • Recipient of full scholarship award from ANU of $94K, Australia, 2016
  • Recipient of travel grant award from ANU of $7K, Australia, 2016
  • Recipient of CSIRO Top-up Award of $35K, Australia, 2016
  • Recipient of NICTA Project grant of $10K, Australia, 2016