Optical flow recording and analysis using Dynamic Vision Sensors
Type and duration:
Bachelor project, flexible duration.
Context:
Dynamic Vision Sensors (DVS) [1] are image sensors that produce so-called events when local changes in brightness are detected. This project will entail using such sensors to record a dataset of controlled motion from which optical flow information will be calculated. The goal will be to design, setup and implement an experimental procedure to collect a dataset that can be analyzed using algorithms such as the Lucas-Kanade Method [2]. This dataset can then be used to test spiking neural network architectures for optical flow estimation with a benchmark.
Objectives:
- Literature review of sensors and methods
- Design experimental setup
- Construct Experimental setup
- Recording of Data
- Analysis of data using algorithms such as the Lucas-Kanade Method
Required skills:
Python programming with Pandas, linear algebra
Contact person:
References:
- P. Lichtsteiner, C. Posch and T. Delbruck, "A 128 × 128 120 dB 15 μ s Latency Asynchronous Temporal Contrast Vision Sensor," in IEEE Journal of Solid-State Circuits, vol. 43, no. 2, pp. 566-576, Feb. 2008, doi: 10.1109/JSSC.2007.914337.
| Last modified: | 15 February 2023 11.21 a.m. |