HOG feature extractor circuit for real-time human and vehicle detection

International audienceIn this paper we propose HOG-Dot, a method for the direct computation of the polar image gradients coordinates from the pixels values. The proposed algorithm, to be used as the first step of the Histogram of Oriented Gradient (HOG) pipeline, approximates the exact gradient with its projection onto a versor chosen among the projection plane set. Instead of non linear computations, the HOG-Dot method exploits linear operations while introducing a bounded approximation error with respect to other HOG approaches, thus resulting a more suitable solution for embedded devices. Concerning the state of the art, it also achieves improved accuracy with the mathematical spatial gradient formulation

show abstract

“…(10), the angle of the vector belongs to the following interval: (11) Thus, the HOG-Dot method approximates the exact gradient as follows:…”

Section: Methodsmentioning

confidence: 99%

HOG-Dot: A Parallel Kernel-Based Gradient Extraction for Embedded Image Processing

Maggiani

Bourrasset

Petracca

et al. 2015

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

show abstract

“…In [18] a HOG feature extractor circuit for pedestrian and vehicle detection, using fixed-point data, was described with an estimated throughput of 33 fps at a single scale for 640 × 480 images. The detection accuracy was not reported or compared to a reference implementation.…”

Section: Background a Related Workmentioning

confidence: 99%

Evaluation and Acceleration of High-Throughput Fixed-Point Object Detection on FPGAs

Najjar

Roy-Chowdhury

2015

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Abstract-The reliance on object or people detection is rapidly growing beyond surveillance to industrial and social applications. The Histogram of Oriented Gradients (HOG), one of the most popular object detection algorithms, achieves high detection accuracy but delivers just under one frame-per-second (fps) on a high-end CPU. FPGA accelerations of this algorithm are limited by the intensive floating-point computations. All current fixedpoint HOG implementations use large bit-width to maintain detection accuracy, or perform poorly at reduced data precision. In this paper we introduce the full-image evaluation methodology to explore the FPGA implementation of HOG using reduced bit-width. This approach lessens the required area resources on the FPGA and increases the clock frequency and hence the throughput per device through increased parallelism. We evaluate the detection accuracy of the fixed-point HOG by applying state-of-the-art computer vision pedestrian detection evaluation metrics and show it performs as well as the original floatingpoint code from OpenCV. We then show our single FPGA implementation achieves a 68.7x higher throughput than a highend CPU, 5.1x higher than a high-end GPU, and 7.8x higher than the same implementation using floating-point on the same FPGA. A power consumption comparison for different platforms shows our fixed-point FPGA implementation uses 130x less power than CPU, and 31x less energy than GPU to process one image.

show abstract

“…Although works either provide only qualitative results [12], or use limited non-public sequences and databases [13] [11], in [14] a large public database of vehicle hypothesis obtained from an on-board forward looking camera is considered. Computational efficiency is addressed differently in the literature: from standard HOG ad-hoc hardware implementations [15], to descriptor simplifications [12] [14] [16] reducing the orientation range considered, modifying the weighted contributions to adjacent bins, or proposing alternative cell and blocks configurations to alleviate the cost of classification. The use of different block sizes is explored in [17], with fairly low accuracy results, while in [18] the use of masks adapted to the vehicle shape is proposed to speed up classification with good results in the classification between different types of vehicles.…”

Section: Previous Workmentioning

confidence: 99%

Histograms of oriented gradients for fast on-board vehicle verification

Ballesteros

Salgado

2014

2014 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Histograms of Oriented Gradients (HoGs) provide excellent results in object detection and verification. However, their demanding processing requirements bound their applicability in some critical real-time scenarios, such as for video-based on-board vehicle detection systems. In this work, an efficient HOG configuration for pose-based on-board vehicle verification is proposed, which alleviates both the processing requirements and required feature vector length without reducing classification performance. The impact on classification of some critical configuration and processing parameters is in depth analyzed to propose a baseline efficient descriptor. Based on the analysis of its cells contribution to classification, new view-dependent cell-configuration patterns are proposed, resulting in reduced descriptors which provide an excellent balance between performance and computational requirements, rendering higher verification rates than other works in the literature.

show abstract

HOG feature extractor circuit for real-time human and vehicle detection

Cited by 13 publications

References 12 publications

HOG-Dot: A Parallel Kernel-Based Gradient Extraction for Embedded Image Processing

HOG-Dot: A Parallel Kernel-Based Gradient Extraction for Embedded Image Processing

Evaluation and Acceleration of High-Throughput Fixed-Point Object Detection on FPGAs

Histograms of oriented gradients for fast on-board vehicle verification

Contact Info

Product

Resources

About