Ryad Benosman

Marco Rasetto, Himanshu Akolkar, Ryad Benosman

The Hierarchy Of Time-Surfaces (HOTS) algorithm, a neuromorphic approach for feature extraction from event data, presents promising capabilities but faces challenges in accuracy and compatibility with neuromorphic hardware. In this paper, we introduce Sup3r, a Semi-Supervised algorithm aimed at addressing these challenges. Sup3r enhances sparsity, stability, and separability in the HOTS networks. It enables end-to-end online training of HOTS networks replacing external classifiers, by leveraging semi-supervised learning. Sup3r learns class-informative patterns, mitigates confounding features, and reduces the number of processed events. Moreover, Sup3r facilitates continual and incremental learning, allowing adaptation to data distribution shifts and learning new tasks without forgetting. Preliminary results on N-MNIST demonstrate that Sup3r achieves comparable accuracy to similarly sized Artificial Neural Networks trained with back-propagation. This work showcases the potential of Sup3r to advance the capabilities of HOTS networks, offering a promising avenue for neuromorphic algorithms in real-world applications.

Daniel C. Stumpp, Himanshu Akolkar, Alan D. George et al.

As the use of neuromorphic, event-based vision sensors expands, the need for compression of their output streams has increased. While their operational principle ensures event streams are spatially sparse, the high temporal resolution of the sensors can result in high data rates from the sensor depending on scene dynamics. For systems operating in communication-bandwidth-constrained and power-constrained environments, it is essential to compress these streams before transmitting them to a remote receiver. Therefore, we introduce a flow-based method for the real-time asynchronous compression of event streams as they are generated. This method leverages real-time optical flow estimates to predict future events without needing to transmit them, therefore, drastically reducing the amount of data transmitted. The flow-based compression introduced is evaluated using a variety of methods including spatiotemporal distance between event streams. The introduced method itself is shown to achieve an average compression ratio of 2.81 on a variety of event-camera datasets with the evaluation configuration used. That compression is achieved with a median temporal error of 0.48 ms and an average spatiotemporal event-stream distance of 3.07. When combined with LZMA compression for non-real-time applications, our method can achieve state-of-the-art average compression ratios ranging from 10.45 to 17.24. Additionally, we demonstrate that the proposed prediction algorithm is capable of performing real-time, low-latency event prediction.

Marco Rasetto, Juan P. Dominguez-Morales, Angel Jimenez-Fernandez et al.

In recent years tremendous efforts have been done to advance the state of the art for Natural Language Processing (NLP) and audio recognition. However, these efforts often translated in increased power consumption and memory requirements for bigger and more complex models. These solutions falls short of the constraints of IoT devices which need low power, low memory efficient computation, and therefore they fail to meet the growing demand of efficient edge computing. Neuromorphic systems have proved to be excellent candidates for low-power low-latency computation in a multitude of applications. For this reason we present a neuromorphic architecture, capable of unsupervised auditory feature recognition. We then validate the network on a subset of Google's Speech Commands dataset.

Himanshu Akolkar, SioHoi Ieng, Ryad Benosman

Optical flow is a crucial component of the feature space for early visual processing of dynamic scenes especially in new applications such as self-driving vehicles, drones and autonomous robots. The dynamic vision sensors are well suited for such applications because of their asynchronous, sparse and temporally precise representation of the visual dynamics. Many algorithms proposed for computing visual flow for these sensors suffer from the aperture problem as the direction of the estimated flow is governed by the curvature of the object rather than the true motion direction. Some methods that do overcome this problem by temporal windowing under-utilize the true precise temporal nature of the dynamic sensors. In this paper, we propose a novel multi-scale plane fitting based visual flow algorithm that is robust to the aperture problem and also computationally fast and efficient. Our algorithm performs well in many scenarios ranging from fixed camera recording simple geometric shapes to real world scenarios such as camera mounted on a moving car and can successfully perform event-by-event motion estimation of objects in the scene to allow for predictions of upto 500 ms i.e. equivalent to 10 to 25 frames with traditional cameras.

Laurent Dardelet, Sio-Hoi Ieng, Ryad Benosman

This paper presents a new event-based method for detecting and tracking features from the output of an event-based camera. Unlike many tracking algorithms from the computer vision community, this process does not aim for particular predefined shapes such as corners. It relies on a dual intertwined iterative continuous -- pure event-based -- estimation of the velocity vector and a bayesian description of the generative feature contours. By projecting along estimated speeds updated for each incoming event it is possible to identify and determine the spatial location and generative contour of the tracked feature while iteratively updating the estimation of the velocity vector. Results on several environments are shown taking into account large variations in terms of luminosity, speed, nature and size of the tracked features. The usage of speed instead of positions allows for a much faster feedback allowing for very fast convergence rates.

Jean-Matthieu Maro, Ryad Benosman

This paper introduces a framework of gesture recognition operating on the output of an event based camera using the computational resources of a mobile phone. We will introduce a new development around the concept of time-surfaces modified and adapted to run on the limited computational resources of a mobile platform. We also introduce a new method to remove dynamically backgrounds that makes full use of the high temporal resolution of event-based cameras. We assess the performances of the framework by operating on several dynamic scenarios in uncontrolled lighting conditions indoors and outdoors. We also introduce a new publicly available event-based dataset for gesture recognition selected through a clinical process to allow human-machine interactions for the visually-impaired and the elderly. We finally report comparisons with prior works that tackled event-based gesture recognition reporting comparable if not superior results if taking into account the limited computational and memory constraints of the used hardware.

Marco Macanovic, Fabian Chersi, Felix Rutard et al.

We introduce in this paper the principle of Deep Temporal Networks that allow to add time to convolutional networks by allowing deep integration principles not only using spatial information but also increasingly large temporal window. The concept can be used for conventional image inputs but also event based data. Although inspired by the architecture of brain that inegrates information over increasingly larger spatial but also temporal scales it can operate on conventional hardware using existing architectures. We introduce preliminary results to show the efficiency of the method. More in-depth results and analysis will be reported soon!

Germain Haessig, Ryad Benosman

This paper introduces an unsupervised compact architecture that can extract features and classify the contents of dynamic scenes from the temporal output of a neuromorphic asynchronous event-based camera. Event-based cameras are clock-less sensors where each pixel asynchronously reports intensity changes encoded in time at the microsecond precision. While this technology is gaining more attention, there is still a lack of methodology and understanding of their temporal properties. This paper introduces an unsupervised time-oriented event-based machine learning algorithm building on the concept of hierarchy of temporal descriptors called time surfaces. In this work we show that the use of sparse coding allows for a very compact yet efficient time-based machine learning that lowers both the computational cost and memory need. We show that we can represent visual scene temporal dynamics with a finite set of elementary time surfaces while providing similar recognition rates as an uncompressed version by storing the most representative time surfaces using clustering techniques. Experiments will illustrate the main optimizations and trade-offs to consider when implementing the method for online continuous vs. offline learning. We report results on the same previously published 36 class character recognition task and a 4 class canonical dynamic card pip task, achieving 100% accuracy on each.

Gregor Lenz, Sio-Hoi Ieng, Ryad Benosman

We present the first purely event-based method for face detection using the high temporal resolution of an event-based camera. We will rely on a new feature that has never been used for such a task that relies on detecting eye blinks. Eye blinks are a unique natural dynamic signature of human faces that is captured well by event-based sensors that rely on relative changes of luminance. Although an eye blink can be captured with conventional cameras, we will show that the dynamics of eye blinks combined with the fact that two eyes act simultaneously allows to derive a robust methodology for face detection at a low computational cost and high temporal resolution. We show that eye blinks have a unique temporal signature over time that can be easily detected by correlating the acquired local activity with a generic temporal model of eye blinks that has been generated from a wide population of users. We furthermore show that once the face is reliably detected it is possible to apply a probabilistic framework to track the spatial position of a face for each incoming event while updating the position of trackers. Results are shown for several indoor and outdoor experiments. We will also release an annotated data set that can be used for future work on the topic.

Amos Sironi, Manuele Brambilla, Nicolas Bourdis et al.

Event-based cameras have recently drawn the attention of the Computer Vision community thanks to their advantages in terms of high temporal resolution, low power consumption and high dynamic range, compared to traditional frame-based cameras. These properties make event-based cameras an ideal choice for autonomous vehicles, robot navigation or UAV vision, among others. However, the accuracy of event-based object classification algorithms, which is of crucial importance for any reliable system working in real-world conditions, is still far behind their frame-based counterparts. Two main reasons for this performance gap are: 1. The lack of effective low-level representations and architectures for event-based object classification and 2. The absence of large real-world event-based datasets. In this paper we address both problems. First, we introduce a novel event-based feature representation together with a new machine learning architecture. Compared to previous approaches, we use local memory units to efficiently leverage past temporal information and build a robust event-based representation. Second, we release the first large real-world event-based dataset for object classification. We compare our method to the state-of-the-art with extensive experiments, showing better classification performance and real-time computation.

Germain Haessig, Andrew Cassidy, Rodrigo Alvarez et al.

This paper describes a fully spike-based neural network for optical flow estimation from Dynamic Vision Sensor data. A low power embedded implementation of the method which combines the Asynchronous Time-based Image Sensor with IBM's TrueNorth Neurosynaptic System is presented. The sensor generates spikes with sub-millisecond resolution in response to scene illumination changes. These spike are processed by a spiking neural network running on TrueNorth with a 1 millisecond resolution to accurately determine the order and time difference of spikes from neighboring pixels, and therefore infer the velocity. The spiking neural network is a variant of the Barlow Levick method for optical flow estimation. The system is evaluated on two recordings for which ground truth motion is available, and achieves an Average Endpoint Error of 11% at an estimated power budget of under 80mW for the sensor and computation.

Garrick Orchard, Cedric Meyer, Ralph Etienne-Cummings et al.

This paper introduces a spiking hierarchical model for object recognition which utilizes the precise timing information inherently present in the output of biologically inspired asynchronous Address Event Representation (AER) vision sensors. The asynchronous nature of these systems frees computation and communication from the rigid predetermined timing enforced by system clocks in conventional systems. Freedom from rigid timing constraints opens the possibility of using true timing to our advantage in computation. We show not only how timing can be used in object recognition, but also how it can in fact simplify computation. Specifically, we rely on a simple temporal-winner-take-all rather than more computationally intensive synchronous operations typically used in biologically inspired neural networks for object recognition. This approach to visual computation represents a major paradigm shift from conventional clocked systems and can find application in other sensory modalities and computational tasks. We showcase effectiveness of the approach by achieving the highest reported accuracy to date (97.5\%$\pm$3.5\%) for a previously published four class card pip recognition task and an accuracy of 84.9\%$\pm$1.9\% for a new more difficult 36 class character recognition task.

Xavier Lagorce, Ryad Benosman

There has been significant research over the past two decades in developing new platforms for spiking neural computation. Current neural computers are primarily developed to mimick biology. They use neural networks which can be trained to perform specific tasks to mainly solve pattern recognition problems. These machines can do more than simulate biology, they allow us to re-think our current paradigm of computation. The ultimate goal is to develop brain inspired general purpose computation architectures that can breach the current bottleneck introduced by the Von Neumann architecture. This work proposes a new framework for such a machine. We show that the use of neuron like units with precise timing representation, synaptic diversity, and temporal delays allows us to set a complete, scalable compact computation framework. The presented framework provides both linear and non linear operations, allowing us to represent and solve any function. We show usability in solving real use cases from simple differential equations to sets of non-linear differential equations leading to chaotic attractors.