arXiv subjects

Xin Lou

Publications and source records attributed to Xin Lou.

28 records · Page 2Linked to original sources

arXiv2023

A Numerical-based Parametric Error Analysis Method for Goldschmidt Floating Point Division

This paper proposes a parametric error analysis method for Goldschmidt floating point division, which reveals how the errors of the intermediate results accumulate and propagate during the Goldschmidt iterations. The analysis is developed by separating the error terms with and without convergence to zero, which are the key parts of the iterative approximate value. The proposed method leads to a state-of-the-art wordlength reduction for intermediate results during the Goldschmidt iterative computation. It enables at least half of the calculation precision reduction for the iterative factor to implement the rectangular multiplier in the divider through flexible numerical method, which can also be applied in the analysis of Goldschmidt iteration with iterative factors assigned under other ways for faster convergence. Based on the proposed method, two proof-of-concept divider models with different configurations are developed, which are verified by more than 100 billion random test vectors to show the correctness and tightness of the proposed error analysis.

math.NA↗

arXiv2022

Learned Smartphone ISP on Mobile GPUs with Deep Learning, Mobile AI & AIM 2022 Challenge: Report

The role of mobile cameras increased dramatically over the past few years, leading to more and more research in automatic image quality enhancement and RAW photo processing. In this Mobile AI challenge, the target was to develop an efficient end-to-end AI-based image signal processing (ISP) pipeline replacing the standard mobile ISPs that can run on modern smartphone GPUs using TensorFlow Lite. The participants were provided with a large-scale Fujifilm UltraISP dataset consisting of thousands of paired photos captured with a normal mobile camera sensor and a professional 102MP medium-format FujiFilm GFX100 camera. The runtime of the resulting models was evaluated on the Snapdragon's 8 Gen 1 GPU that provides excellent acceleration results for the majority of common deep learning ops. The proposed solutions are compatible with all recent mobile GPUs, being able to process Full HD photos in less than 20-50 milliseconds while achieving high fidelity results. A detailed description of all models developed in this challenge is provided in this paper.

cs.CV↗

arXiv2022

ICARUS: A Specialized Architecture for Neural Radiance Fields Rendering

The practical deployment of Neural Radiance Fields (NeRF) in rendering applications faces several challenges, with the most critical one being low rendering speed on even high-end graphic processing units (GPUs). In this paper, we present ICARUS, a specialized accelerator architecture tailored for NeRF rendering. Unlike GPUs using general purpose computing and memory architectures for NeRF, ICARUS executes the complete NeRF pipeline using dedicated plenoptic cores (PLCore) consisting of a positional encoding unit (PEU), a multi-layer perceptron (MLP) engine, and a volume rendering unit (VRU). A PLCore takes in positions \& directions and renders the corresponding pixel colors without any intermediate data going off-chip for temporary storage and exchange, which can be time and power consuming. To implement the most expensive component of NeRF, i.e., the MLP, we transform the fully connected operations to approximated reconfigurable multiple constant multiplications (MCMs), where common subexpressions are shared across different multiplications to improve the computation efficiency. We build a prototype ICARUS using Synopsys HAPS-80 S104, a field programmable gate array (FPGA)-based prototyping system for large-scale integrated circuits and systems design. We evaluate the power-performance-area (PPA) of a PLCore using 40nm LP CMOS technology. Working at 400 MHz, a single PLCore occupies 16.5 $mm^2$ and consumes 282.8 mW, translating to 0.105 uJ/sample. The results are compared with those of GPU and tensor processing unit (TPU) implementations.

cs.AR↗

arXiv2022

Hessian filter-assisted full diameter at half maximum (FDHM) segmentation and quantification method for optical-resolution photoacoustic microscopy

Optical-resolution photoacoustic microscopy has been validated as a high-resolution and high-sensitivity imaging modality for angiographic studies in the past decades. Quantitative vascular analysis reveals critical information of physiological changes, where vessel segmentation is the key step. In this work, we developed a Hessian filter-assisted, adaptive thresholding vessel segmentation algorithm. Its performance is validated by a digital phantom and in vivo images. Its capability of capturing subtle vessel changes is further tested in two longitudinal studies on vascular responses to blood pressure agents. The results are compared with the widely used Hessian filter method. In the antihypotensive case, the proposed method detected a twice larger vasoconstriction than the Hessian filter method. In the antihypertensive case, the proposed method detected a vasodilation of 18.8 %, while the Hessian filter method failed in change detection. The proposed algorithm could correct errors caused by conventional segmentation methods and improve quantitative accuracy for angiographic applications.

physics.med-ph↗

arXiv2021

Raw Bayer Pattern Image Synthesis for Computer Vision-oriented Image Signal Processing Pipeline Design

In this paper, we propose a method to add constraints that are un-formulatable in generative adversarial networks (GAN)-based arbitrary size RAW Bayer image generation. It is shown theoretically that by using the transformed data in GAN training, it is able to improve the learning of the original data distribution, owing to the invariant of Jensen-Shannon (JS) divergence between two distributions under invertible and differentiable transformation. Benefiting from the proposed method, RAW Bayer pattern images can be generated by configuring the transformation as demosaicing. It is shown that by adding another transformation, the proposed method is able to synthesize high-quality RAW Bayer images with arbitrary size. Experimental results show that images generated by the proposed method outperform the existing methods in the Fréchet inception distance (FID) score, peak signal to noise ratio (PSNR), and mean structural similarity (MSSIM), and the training process is more stable. To the best knowledge of the authors, there is no open-source, large-scale image dataset in the RAW Bayer domain, which is crucial for research works aiming to explore the image signal processing (ISP) pipeline design for computer vision tasks. Converting the existing commonly used color image datasets to their corresponding RAW Bayer versions, the proposed method can be a promising solution to the RAW image dataset problem. We also show in the experiments that, by training object detection frameworks using the synthesized RAW Bayer images, they can be used in an end-to-end manner (from RAW images to vision tasks) with negligible performance degradation.

eess.IV↗

arXiv2021

Gradient-based Feature Extraction From Raw Bayer Pattern Images

In this paper, the impact of demosaicing on gradient extraction is studied and a gradient-based feature extraction pipeline based on raw Bayer pattern images is proposed. It is shown both theoretically and experimentally that the Bayer pattern images are applicable to the central difference gradient-based feature extraction algorithms without performance degradation, or even with superior performance in some cases. The color difference constancy assumption, which is widely used in various demosaicing algorithms, is applied in the proposed Bayer pattern image-based gradient extraction pipeline. Experimental results show that the gradients extracted from Bayer pattern images are robust enough to be used in histogram of oriented gradients (HOG)-based pedestrian detection algorithms and shift-invariant feature transform (SIFT)-based matching algorithms.

eess.IV↗

arXiv2021

Compressing Large-Scale Transformer-Based Models: A Case Study on BERT

Pre-trained Transformer-based models have achieved state-of-the-art performance for various Natural Language Processing (NLP) tasks. However, these models often have billions of parameters, and, thus, are too resource-hungry and computation-intensive to suit low-capability devices or applications with strict latency requirements. One potential remedy for this is model compression, which has attracted a lot of research attention. Here, we summarize the research in compressing Transformers, focusing on the especially popular BERT model. In particular, we survey the state of the art in compression for BERT, we clarify the current best practices for compressing large-scale Transformer models, and we provide insights into the workings of various methods. Our categorization and analysis also shed light on promising future research directions for achieving lightweight, accurate, and generic NLP models.

cs.LG↗

arXiv2020

On Lightweight Privacy-Preserving Collaborative Learning for Internet of Things by Independent Random Projections

The Internet of Things (IoT) will be a main data generation infrastructure for achieving better system intelligence. This paper considers the design and implementation of a practical privacy-preserving collaborative learning scheme, in which a curious learning coordinator trains a better machine learning model based on the data samples contributed by a number of IoT objects, while the confidentiality of the raw forms of the training data is protected against the coordinator. Existing distributed machine learning and data encryption approaches incur significant computation and communication overhead, rendering them ill-suited for resource-constrained IoT objects. We study an approach that applies independent random projection at each IoT object to obfuscate data and trains a deep neural network at the coordinator based on the projected data from the IoT objects. This approach introduces light computation overhead to the IoT objects and moves most workload to the coordinator that can have sufficient computing resources. Although the independent projections performed by the IoT objects address the potential collusion between the curious coordinator and some compromised IoT objects, they significantly increase the complexity of the projected data. In this paper, we leverage the superior learning capability of deep learning in capturing sophisticated patterns to maintain good learning performance. The extensive comparative evaluation shows that this approach outperforms other lightweight approaches that apply additive noisification for differential privacy and/or support vector machines for learning in the applications with light to moderate data pattern complexities.

cs.CR↗

arXiv2019

On Lightweight Privacy-Preserving Collaborative Learning for IoT Objects

The Internet of Things (IoT) will be a main data generation infrastructure for achieving better system intelligence. This paper considers the design and implementation of a practical privacy-preserving collaborative learning scheme, in which a curious learning coordinator trains a better machine learning model based on the data samples contributed by a number of IoT objects, while the confidentiality of the raw forms of the training data is protected against the coordinator. Existing distributed machine learning and data encryption approaches incur significant computation and communication overhead, rendering them ill-suited for resource-constrained IoT objects. We study an approach that applies independent Gaussian random projection at each IoT object to obfuscate data and trains a deep neural network at the coordinator based on the projected data from the IoT objects. This approach introduces light computation overhead to the IoT objects and moves most workload to the coordinator that can have sufficient computing resources. Although the independent projections performed by the IoT objects address the potential collusion between the curious coordinator and some compromised IoT objects, they significantly increase the complexity of the projected data. In this paper, we leverage the superior learning capability of deep learning in capturing sophisticated patterns to maintain good learning performance. Extensive comparative evaluation shows that this approach outperforms other lightweight approaches that apply additive noisification for differential privacy and/or support vector machines for learning in the applications with light data pattern complexities.

cs.LG↗

arXiv2017

One-Hop Out-of-Band Control Planes for Low-Power Multi-Hop Wireless Networks

Separation of control and data planes (SCDP) is a desirable paradigm for low-power multi-hop wireless networks requiring high network performance and manageability. Existing SCDP networks generally adopt an in-band control plane scheme in that the control-plane messages are delivered by their data-plane networks. The physical coupling of the two planes may lead to undesirable consequences. To advance the network architecture design, we propose to leverage on the long-range communication capability of the increasingly available low-power wide-area network (LPWAN) radios to form one-hop out-of-band control planes. We choose LoRaWAN, an open, inexpensive, and ISM band based LPWAN radio to prototype our out-of-band control plane called LoRaCP. Several characteristics of LoRaWAN such as downlink-uplink asymmetry and primitive ALOHA media access control (MAC) present challenges to achieving reliability and efficiency. To address these challenges, we design a TDMA-based multi-channel MAC featuring an urgent channel and negative acknowledgment. On a testbed of 16 nodes, we demonstrate applying LoRaCP to physically separate the control-plane network of the Collection Tree Protocol (CTP) from its ZigBee-based data-plane network. Extensive experiments show that LoRaCP increases CTP's packet delivery ratio from 65% to 80% in the presence of external interference, while consuming a per-node average radio power of 2.97mW only.

cs.NI↗