Aryan Esfandiari
Principal Machine Learning Engineer and Researcher
Biography
Interests
- Artificial Intelligence
- Neural Networks
- Deep Learning
- Computer Vision
- Generative Models
- Natural Language Processing
- Multimodal LLMs
- Robotics
Education
- Doctor of Philosophy in Medical Artificial Intelligence, 2022King’s and Imperial College London
- Master of Science in Robotics and Intelligent Systems, 2016University of Oslo and University of California of Berkeley
- Bachelor of Science in Robotics and Intelligent Systems, 2014University of Oslo
Skills
Artificial Intelligence
Computer Vision
Robotics
Experience
Principal Artificial Intelligence Researcher and Engineer
Huawei Technologies R&D UKJan 2019 – Present · Cambridge, United Kingdom
Cambridge and London Research Centre (2012 Laboratories) – Computer Vision
Research on the following topics:
- Supervised and self-supervised learning
- Convolutional neural networks
- Transformers
- Generative and adversarial networks
- Autoencoders and recurrent neural networks
- Model compression, pruning, quantization and knowledge distillation
Senior Machine Learning Design Engineer
Advanced Micro DevicesOct 2016 – Jan 2019 · Edinburgh, United Kingdom
- Design of artificial intelligence accelerators
- Implementation of efficient deep convolutional neural networks
- Development of deep learning architechtures, frameworks and libraries
- Model optimisation, compression and quantization
- On-device deployment for time-critical and real-time applications
- Teach both undergraduate and postgraduate level courses
- Supervision of students with dissertations and publications
- Development of curriculum and teaching materials
- Development of advanced embedding systems and Internet of Things (IoT)
- On-board data acquisition and processing
- Digital electronics
- Computer-aided design
AIM 2020 Challenge on Video Extreme Super-Resolution: Methods and Results
This paper reviews the video extreme super-resolution challenge associated with the AIM 2020 workshop at ECCV 2020. Common scaling factors for learned video super-resolution (VSR) do not go beyond factor 4. Missing information can be restored well in this region, especially in HR videos, where the high-frequency content mostly consists of texture details. The task in this challenge is to upscale videos with an extreme factor of 16, which results in more serious degradations that also affect the structural integrity of the videos. A single pixel in the low-resolution (LR) domain corresponds to 256 pixels in the high-resolution (HR) domain. Due to this massive information loss, it is hard to accurately restore the missing information. Track 1 is set up to gauge the state-of-the-art for such a demanding task, where fidelity to the ground truth is measured by PSNR and SSIM. Perceptually higher quality can be achieved in trade-off for fidelity by generating plausible high-frequency content. Track 2 therefore aims at generating visually pleasing results, which are ranked according to human perception, evaluated by a user study. In contrast to single image super-resolution (SISR), VSR can benefit from additional information in the temporal domain. However, this also imposes an additional requirement, as the generated frames need to be consistent along time.