Req-Yolo

Abstract

Deep neural networks (DNNs), as the basis of object detection, will play a key role in the development of future autonomous systems with full autonomy. The autonomous systems have special requirements of real-time, energy-efficient implementations of DNNs on a power-constrained system. Two research thrusts are dedicated to performance and energy efficiency enhancement of the inference phase of DNNs. The first one is model compression techniques while the second is efficient hardware implementation. Recent works on extremely-low-bit CNNs such as the binary neural network (BNN) and XNOR-Net replace the traditional floating point operations with binary bit operations which significantly reduces the memory bandwidth and storage requirement. However, it suffers from nonnegligible accuracy loss and underutilized digital signal processing (DSP) blocks of FPGAs.To overcome these limitations, this paper proposes REQ-YOLO, a resource aware, systematic weight quantization framework for object detection, considering both algorithm and hardware resource aspects in object detection. We adopt the block-circulant matrix method and propose a heterogeneous weight quantization using Alternating Direction Method of Multipliers (ADMM), an effective optimization technique for general, non-convex optimization problems. To achieve real-time, highly-efficient implementations on FPGA, we present the detailed hardware implementation of block circulant matrices on CONV layers and develop an efficient processing element (PE) structure supporting the heterogeneous weight quantization, CONV dataflow and pipelining techniques, design optimization, and a template-based automatic synthesis framework to optimally exploit hardware resource. Experimental results show that our proposed REQ-YOLO framework can significantly compress the YOLO model while introducing very small accuracy degradation.

Augmentation-Assisted Robust Fringe Detection on Unseen Experimental Signals Applied to Optical Feedback Interferometry Using a Deep Network

“…Is it possible to further optimize this detector for our case of binary images? One popular technique for optimizing 2D deep networks is to quantize floating point operations to integer operations [46], [47]. The weights of 2D deep networks are generally stored as 32-bit floating point numbers (float32).…”

“…Converting these weights to 16-bit floating point numbers (float16) or 8-bit integers (int8) provides memory efficiency, reduces inference latency, and allows leveraging embedded platforms [46]. However, this is achieved at the cost of reduced performance [47].…”

Augmentation assisted robust fringe detection on unseen experimental signals applied to optical feedback interferometry using a deep network

“…this detector for our case of binary images? One popular technique for optimizing 2D deep networks is to quantize floating point operations to integer operations [46], [47]. The weights of 2D deep networks are generally stored as 32-bit floating point numbers (float32).…”

AddNet: Deep Neural Networks Using FPGA-Optimized Multipliers

“…There have been several accelerator architecture designs for low-precision CNNs with uniform quantization arithmetic. Recent literature includes commercial architectures [22], [23] and also academic approaches [24]- [28]. The benefits, in terms of power and throughput, of fitting a design on-chip was Example of a reconfigurable multiplier with the coefficient set {12305, 20746}.…”