Real-time systems

Birch: A Real-Time Multi-Domain Multi-Task Extended Reality Perception Accelerator

Birch: A Real-Time Multi-Domain Multi-Task Extended Reality Perception Accelerator 150 150

Abstract:

Birch is a system-on-chip (SoC) that efficiently and accurately accelerates the multi-task multi-domain extended reality (XR) perception pipeline, with workloads such as visual inertial odometry (VIO), eye gaze tracking, and scene understanding. Birch features vision modules with cascaded line buffers, in-step feature sorting, and double-buffered optical flow to extract and …

View on IEEE Xplore

A 65-nm CMOS Machine-Learning-Enhanced Bandwidth-Efficient LiDAR

A 65-nm CMOS Machine-Learning-Enhanced Bandwidth-Efficient LiDAR 150 150

Abstract:

We present a proof-of-concept light detection and ranging (LiDAR) signal processing architecture that integrates a machine-learning-enhanced processing unit (PU) with on-chip time-to-digital converters (TDCs) to reduce bandwidth and memory requirements in SPAD-based direct time-of-flight (dToF) systems. The proposed architecture fits a Gaussian mixture model (GMM) to photon arrival time distributions …

View on IEEE Xplore

Actiniaria: Distributed Dynamic-IR-Drop-Aware Timing Monitor for AVFS With Lightweight Tentacles

Actiniaria: Distributed Dynamic-IR-Drop-Aware Timing Monitor for AVFS With Lightweight Tentacles 150 150

Abstract:

Advances in integrated circuit (IC) technology have amplified the effects of process, voltage, and temperature (PVT) variations, particularly dynamic IR drop, which severely affects timing. Post-silicon IR drop monitoring circuits are lacking, forcing designers to reserve substantial static guard bands for worst-case scenarios, compromising energy efficiency. Inspired by biomimetics, this …

View on IEEE Xplore

A 57.3-fps 12.8 TFLOPS/W Text-to-Motion Processor With Inter-Iteration Output Sparsity and Inter-Frame Joint Similarity

A 57.3-fps 12.8 TFLOPS/W Text-to-Motion Processor With Inter-Iteration Output Sparsity and Inter-Frame Joint Similarity 150 150

Abstract:

Recently, 3-D human motion generation has become essential in media applications such as film production and augmented reality (AR)/virtual reality (VR) devices, requiring the generation of human joint movements and detailed 3-D meshes for each joint. Traditionally, joint creation required hours or even days, making it impractical for real-time …

View on IEEE Xplore

An Energy-Efficient CNN Processor Supporting Bi-Directional FPN for Small-Object Detection on High-Resolution Videos in 16-nm FinFET

An Energy-Efficient CNN Processor Supporting Bi-Directional FPN for Small-Object Detection on High-Resolution Videos in 16-nm FinFET 150 150

Abstract:

The capability to detect small objects precisely in real time is essential for intelligent systems, particularly in advanced driver assistance systems (ADASs), as it ensures continuous awareness of distant obstacles for enhanced safety. However, achieving high detection precision for small objects requires high-resolution input inference on deep convolutional neural network (…

View on IEEE Xplore

PANNA: A 558 TOPS/W Pipelined All-Analog Neural Network Accelerator in 22 nm FD-SOI

PANNA: A 558 TOPS/W Pipelined All-Analog Neural Network Accelerator in 22 nm FD-SOI 150 150

Abstract:

Analog computing offers intrinsic energy and latency benefits that makes it attractive for real-time and edge applications. Conventional analog accelerators suffer from repeated conversions between analog and digital domain, which degrades efficiency and throughput. We propose an all-analog pipelined neural network accelerator architecture in 22 nm fully-depleted silicon-on-insulator (FD-SOI) complementary metal-oxide-semiconductor (…

View on IEEE Xplore

Space-Mate: A 303.5-mW Real-Time Sparse Mixture-of-Experts-Based NeRF-SLAM Processor for Mobile Spatial Computing

Space-Mate: A 303.5-mW Real-Time Sparse Mixture-of-Experts-Based NeRF-SLAM Processor for Mobile Spatial Computing 150 150

Abstract:

Simultaneous localization and mapping (SLAM) provides crucial ego-pose information and 3-D maps of the user environment, which are fundamental to emerging mobile spatial computing devices. Dense 3-D mapping and accurate pose estimation are particularly necessary for applications like augmented reality (AR) and autonomous navigation. However, existing SLAM processors are typically …

View on IEEE Xplore

An Area and Power Efficient Source Driver IC With Multi-Line Sensing Real-Time Pixel Compensation for OLED Displays

An Area and Power Efficient Source Driver IC With Multi-Line Sensing Real-Time Pixel Compensation for OLED Displays 150 150

Abstract:

This article proposes a source driver integrated circuit (SD-IC) with a pixel compensator designed to minimize both silicon area and power consumption for organic light-emitting diode (OLED) displays. The proposed SD-IC is capable of driving ultrahigh-definition (UHD) active-matrix OLED (AMOLED) panels with a one-horizontal-time (1-H time) of $7.2~\mu $ s while …

View on IEEE Xplore

IRIS: An Energy-Efficient Spatial Computing SoC for Real-Time Interactive Rendering and Modeling With Surface-Aware 3-D Gaussian Splatting

IRIS: An Energy-Efficient Spatial Computing SoC for Real-Time Interactive Rendering and Modeling With Surface-Aware 3-D Gaussian Splatting 150 150

Abstract:

The 3-D Gaussian splatting (3DGS), based on a machine learning-driven radiance field technique, is rapidly emerging as a next-generation solution in 3-D graphics. Owing to its short modeling time, computational simplicity, and high rendering quality, it is expected to replace traditional 3-D graphics on edge devices. However, its substantial memory …

View on IEEE Xplore