Neuromorphic Computer Vision: Sensing and Neuromorphic Machine Learning for Vision Applications

About the Project

Recent advances in bio-inspired neuromorphic hardware and neuromorphic sensors enable more efficient methodologies for computer vision tasks, aiming to minimise the cost, latency and energy consumption of real-time vision systems.

This project will focus on processing event streams generated from neuromorphic cameras and applying neuromorphic machine learning methods such as Spiking Neural Networks (SNNs) for vision tasks such as object segmentation and recognition, human motion analysis and video understanding.

Candidates should have appropriate academic qualifications (first or upper second class honours or MSc degree), in Computer Science, Engineering, Mathematics, Physics or other relevant area, strong background in programming and expertise in Machine Learning / Artificial Intellifence.

Qualified applicants are encouraged to contact Prof Dimitrios Makris (d.makris@kingston.ac.uk) to informally discuss the project.

Supervisor’s profile: https://www.kingston.ac.uk/staff/profile/professor-dimitrios-makris-151/

Google Scholar profile: https://scholar.google.co.uk/citations?user=vHv7JRcAAAAJ

Funding Notes

there is no funding for this project

References

[1] Kachole, Sanket, Sajwani, Hussain, Baghaei Naeini, Fariborz, Makris, Dimitrios and Zweiri, Yahya (2024) Asynchronous bioplausible neuron for spiking neural networks for event-based vision. In: European Conference on Computer Vision (ECCV) 2024; 29 Sep - 04 Oct 2024, Milan, Italy.
[2] Huang, Xiaoqian, Kachole, Sanket, Ayyad, Abdulla, Baghaei Naeini, Fariborz, Makris, Dimitrios and Zweiri, Yahya (2024) A neuromorphic dataset for tabletop object segmentation in indoor cluttered environment. Scientific Data, 11(127), ISSN (online) 2052-4463
[3] Kachole, Sanket, Huang, Xiaoquan, Baghaei Naeini, Fariborz, Muthusamy, Rajkumar, Makris, Dimitrios and Zweiri, Yahya (2023) Bimodal SegNet : fused instance segmentation using events and RGB frames. Pattern Recognition, 149, p. 110215. ISSN (print) 0031-3203
[4] Kachole, Sanket, Alkendi, Yusra, Baghaei Naeini, Fariborz, Makris, Dimitrios and Zweiri, Yahya (2023) Asynchronous events-based panoptic segmentation using graph mixer neural network. In: 4th International Workshop on Event-Based Vision; 17-24 Jun 2023, Vancouver, Canada.
[5] Baghaei Naeini, Fariborz, Kachole, Sanket, Muthusamy, Rajkumar, Makris, Dimitrios and Zweiri, Yahya (2022) Event augmentation for contact force measurements. IEEE Access, 10, pp. 123651-123660. ISSN (online) 2169-3536
[6] Baghaei Naeini, Fariborz, Makris, Dimitrios, Gan, Dongming and Zweiri, Yahya (2020) Dynamic-vision-based force measurements using convolutional recurrent neural networks. Sensors, 20(16), p. 4469. ISSN (online) 1424-8220
[7] Baghaei Naeini, Fariborz, Alali, Aamna, Al-Husari, Raghad, Rigi, Amin, AlSharman, Mohammad K., Makris, Dimitrios and Zweiri, Yahya (2020) A novel dynamic-vision-based approach for tactile sensing applications. IEEE Transactions on Instrumentation and Measurement, 69(5), pp. 1881-1893. ISSN (print) 0018-9456
[8] Rigi, Amin, Naeini, Fariborz Baghaei, Makris, Dimitrios and Zweiri, Yahya (2018) A novel event-based incipient slip detection using Dynamic Active-Pixel Vision Sensor (DAVIS). Sensors, 18(2), p. 333. ISSN (online) 1424-8220