An official website of the United States government
Convolutional Neural Networks (CNNs), due to their recent successes, have gained lots of attention in various vision-based applications. They have proven to produce incredible results, especially on big data, that require high processing demands. However, CNN processing demands have limited their usage in embedded edge devices with constrained energy budgets and hardware. This paper proposes an efficient new architecture, namely Ocelli includes a ternary compute pixel (TCP) consisting of a CMOS-based pixel and a compute add-on. The proposed Ocelli architecture offers several features; (I) Because of the compute add-on, TCPs can produce ternary values (i.e., −1, 0, +1) regarding the light intensity as pixels’ inputs; (II) Ocelli realizes analog convolutions enabling low-precision ternary weight neural networks. Since the first layer’s convolution operations are the performance bottleneck of accelerators, Ocelli mitigates the overhead of analog buffers and analog-to-digital converters. Moreover, our design supports a zero-skipping scheme to further power reduction; (III) Ocelli exploits non-volatile magnetic RAMs to store CNN’s weights, which remarkably reduces the static power consumption; and finally, (IV) Ocelli has two modes, including sensing and processing. Once the object is detected, the architecture switches to the typical sensing mode to capture the image. Compared to the conventional pixels, it achieves an average 10% efficiency on its lane detection power consumption compared with existing edge detection algorithms. Moreover, considering different CNN workloads, our design shows more than 23% power efficiency over conventional designs, while it can achieve better accuracy.
more »
« less
TizBin: A Low-Power Image Sensor with Event and Object Detection Using Efficient Processing-in-Pixel Schemes
In the Artificial Intelligence of Things (AIoT) era, always-on intelligent and self-powered visual perception systems have gained considerable attention and are widely used. Thus, this paper proposes TizBin, a low-power processing in-sensor scheme with event and object detection capabilities to eliminate power costs of data conversion and transmission and enable data-intensive neural network tasks. Once the moving object is detected, TizBin architecture switches to the high-power object detection mode to capture the image. TizBin offers several unique features, such as analog convolutions enabling low-precision ternary weight neural networks (TWNN) to mitigate the overhead of analog buffer and analog-to-digital converters. Moreover, TizBin exploits non-volatile magnetic RAMs to store NN’s weights, remarkably reducing static power consumption. Our circuit-to-application co-simulation results for TWNNs demonstrate minor accuracy degradation on various image datasets, while TizBin achieves a frame rate of 1000 and efficiency of ∼1.83 TOp/s/W.
more »
« less
- PAR ID:
- 10426811
- Date Published:
- Journal Name:
- 2022 IEEE 40th International Conference on Computer Design (ICCD)
- Page Range / eLocation ID:
- 770 to 777
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Recent advances in computer vision has led to a growth of interest in deploying visual analytics model on mobile devices. However, most mobile devices have limited computing power, which prohibits them from running large scale visual analytics neural networks. An emerging approach to solve this problem is to offload the computation of these neural networks to computing resources at an edge server. Efficient computation offloading requires optimizing the trade-off between multiple objectives including compressed data rate, analytics performance, and computation speed. In this work, we consider a “split computation” system to offload a part of the computation of the YOLO object detection model. We propose a learnable feature compression approach to compress the intermediate YOLO features with lightweight computation. We train the feature compression and decompression module together with the YOLO model to optimize the object detection accuracy under a rate constraint. Compared to baseline methods that apply either standard image compression or learned image compression at the mobile and perform image de-compression and YOLO at the edge, the proposed system achieves higher detection accuracy at the low to medium rate range. Furthermore, the proposed system requires substantially lower computation time on the mobile device with CPU only.more » « less
-
Automatic video analysis tools are an indispensable component in imaging applications. Object detection, the first and the most important step for automatic video analysis, is implemented in many embedded cameras. The accuracy of object detection relies on the quality of images that are processed. This paper proposes a new image quality model for predicting the performance of object detection on embedded cameras. A video data set is constructed that considers different factors for quality degradation in the imaging process, such as reduced resolution, noise, and blur. The performances of commonly used low-complexity object detection algorithms are obtained for the data set. A no-reference regression model based on a bagging ensemble of regression trees is built to predict the accuracy of object detection using observable features in an image. Experimental results show that the proposed model provides more accurate predictions of image quality for object detection than commonly known image quality measures.more » « less
-
null (Ed.)Low-resolution analog-to-digital converters (ADCs) have been considered as a practical and promising solution for reducing cost and power consumption in massive Multiple-Input-Multiple-Output (MIMO) systems. Unfortunately, low-resolution ADCs significantly distort the received signals, and thus make data detection much more challenging. In this paper, we develop a new deep neural network (DNN) framework for efficient and low-complexity data detection in low-resolution massive MIMO systems. Based on reformulated maximum likelihood detection problems, we propose two model-driven DNN-based detectors, namely OBMNet and FBMNet, for one-bit and few-bit massive MIMO systems, respectively. The proposed OBMNet and FBMNet detectors have unique and simple structures designed for low-resolution MIMO receivers and thus can be efficiently trained and implemented. Numerical results also show that OBMNet and FBMNet significantly outperform existing detection methods.more » « less
-
Recently, Intelligent IoT (IIoT), including various sensors, has gained significant attention due to its capability of sensing, deciding, and acting by leveraging artificial neural networks (ANN). Nevertheless, to achieve acceptable accuracy and high performance in visual systems, a power-delay-efficient architecture is required. In this paper, we propose an ultra-low-power processing in-sensor architecture, namely SenTer, realizing low-precision ternary multi-layer perceptron networks, which can operate in detection and classification modes. Moreover, SenTer supports two activation functions based on user needs and the desired accuracy-energy trade-off. SenTer is capable of performing all the required computations for the MLP's first layer in the analog domain and then submitting its results to a co-processor. Therefore, SenTer significantly reduces the overhead of analog buffers, data conversion, and transmission power consumption by using only one ADC. Additionally, our simulation results demonstrate acceptable accuracy on various datasets compared to the full precision models.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICCD56317.2022.00117
