Event-Based Optical Flow Estimation with Spatio-Temporal Backpropagation Trained Spiking Neural Network.

The advantages of an event camera, such as low power consumption, large dynamic range, and low data redundancy, enable it to shine in extreme environments where traditional image sensors are not competent, especially in high-speed moving target capture and extreme lighting conditions. Optical flow reflects the target's movement information, and the target's detailed movement can be obtained using the event camera's optical flow information. However, the existing neural network methods for optical flow prediction of event cameras has the problems of extensive computation and high energy consumption in hardware implementation. The spike neural network has spatiotemporal coding characteristics, so it can be compatible with the spatiotemporal data of an event camera. Moreover, the sparse coding characteristic of the spike neural network makes it run with ultra-low power consumption on neuromorphic hardware. However, because of the algorithmic and training complexity, the spike neural network has not been applied in the prediction of the optical flow for the event camera. For this case, this paper proposes an end-to-end spike neural network to predict the optical flow of the discrete spatiotemporal data stream for the event camera. The network is trained with the spatio-temporal backpropagation method in a self-supervised way, which fully combines the spatiotemporal characteristics of the event camera while improving the network performance. Compared with the existing methods on the public dataset, the experimental results show that the method proposed in this paper is equivalent to the best existing methods in terms of optical flow prediction accuracy, and it can save 99% more power consumption than the existing algorithm, which is greatly beneficial to the hardware implementation of the event camera optical flow prediction., laying the groundwork for future low-power hardware implementation of optical flow prediction for event cameras.

Adaptive Slicing Method of the Spatiotemporal Event Stream Obtained from a Dynamic Vision Sensor.

The dynamic vision sensor (DVS) measures asynchronously change of brightness per pixel, then outputs an asynchronous and discrete stream of spatiotemporal event information that encodes the time, location, and sign of brightness changes. The dynamic vision sensor has outstanding properties compared to sensors of traditional cameras, with very high dynamic range, high temporal resolution, low power consumption, and does not suffer from motion blur. Hence, dynamic vision sensors have considerable potential for computer vision in scenarios that are challenging for traditional cameras. However, the spatiotemporal event stream has low visualization and is incompatible with existing image processing algorithms. In order to solve this problem, this paper proposes a new adaptive slicing method for the spatiotemporal event stream. The resulting slices of the spatiotemporal event stream contain complete object information, with no motion blur. The slices can be processed either with event-based algorithms or by constructing slices into virtual frames and processing them with traditional image processing algorithms. We tested our slicing method using public as well as our own data sets. The difference between the object information entropy of the slice and the ideal object information entropy is less than 1%.