Supervised learning and low-power processing of data from event based cameras such as dynamic vision sensor (DVS), Asynchronous Time-based Image Sensor (ATIS) etc., is another major focus of research in our team. We are looking to use reconfigurable hardware, such as FPGAs, and ARM processors, for implementation of novel event-based algorithms for power efficient computing.
The Xilinx Zynq-7020 FPGA was interfaced to a down-looking DAVIS camera, on-board an unmanned aerial vehicle, recognizing different objects beneath it.
Besides low-power hardware implementations, we also carry out academically oriented work for object tracking, learning and detection.
Spiking Neural Networks
Spiking Neural Networks use biologically plausible models of neuron as their computational unit. It is one of the major foucus of our research, especially supervised learning for processing data from event based senosrs such as silicon retina, silicon cochlea etc. We are also looking to configure spiking neural networks for implementation in neuromorphic hardwares for power efficient computing.
Bharath Ramesh*, Andrés Ussa, Luca Della Vedova, Hong Yang and Garrick Orchard., "Low-power Dynamic Object Detection and Classification with Freely Moving Event Cameras" Frontiers in Neuroscience (Accepted), Feb. 2020 link
Ramesh B., Yang H., Orchard G., Le Thi N., Zhang S. and Xiang C. "DART: Distribution Aware Retinal Transform for Event-based Cameras" IEEE Transactions on Pattern Analysis and Machine Intelligence, DOI: 10.1109/TPAMI.2019.2919301 , May 2019 link
Zhen X., Sheng Y.C. and Orchard, G. "Event-based Stereo Depth Estimation Using Belief Propagation" Frontiers in Neuroscience vol. 11, pp. 535, Oct. 2017 link
Deepak Singla, Soham Chatterjee, Lavanya Ramapantulu, Andres Ussa, Bharath Ramesh, and Arindam Basu, "HyNNA: Improved Performance for Neuromorphic Vision Sensor Based Surveillance Using Hybrid Neural Network Architecture", IEEE International Symposium on Circuits and Systems (ISCAS) , Sevilla, Spain, May 2020 (Accepted)
Andres Ussa, Luca Della Vedova, Vandana Reddy Padala, Deepak Singla, Jyotibdha Acharya, Charles Zhang Lei, Garrick Orchard, Arindam Basu, Bharath Ramesh., "A low-power end-to-end hybrid neuromorphic framework for surveillance applications", British Machine Vision Conference Workshops, Cardiff, United Kingdom, Sep 2019 PDF, video
Jyotibdha Acharya, Andres Ussa Caycedo, Vandana Reddy Padala, Rishi Raj Sidhu Singh, Garrick Orchard, Bharath Ramesh, Arindam Basu., "EBBIOT: A Low-complexity Tracking Algorithm for Surveillance in IoVT Using Stationary Neuromorphic Vision Sensors", IEEE International System-on-Chip Conference (SOCC), Singapore, Sep 2019 PDF, video
Ramesh B., Ussa Caycedo A. C., Della Vedova L., Yang H. and Orchard G. "PCA-RECT: An Energy-efficient Object Detection Approach for Event Cameras", Asian Conference on Computer Vision, Perth, Australia, Dec 2018
Colonnier F., Della Vedova L., Teo R.S.H and Orchard G. "Obstacle Avoidance using Eventbased Visual Sensor and Time-To-Contact Processing", Australasian Conference on Robotics and Automation, Lincoln, New Zealand, Dec 2018 PDF, video
Tun Aung M., Teo R. and Orchard G. “Event-based Plane-fitting Optical Flow for Dynamic Vision Sensors in FPGA” IEEE Int. Symp. Circuits Syst., Florence, Italy, May 2018
Ramesh B., Le Thi N., Orchard G. and Xiang C. "Spike Context: A Neuromorphic Descriptor for Pattern Recognition", IEEE Biomed. Circuits Syst., Turin, Italy, Oct 2017
Czech D. and Orchard G. "Evaluating Noise Filtering for Event-based Asynchronous Change Detection Image Sensors" 6th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), Singapore, June 2016