Neural networks - IoU & NMS

IoU IoU (Intersection over Union), also known as overlap / intersection union ratio. That is, the intersection/Union in the figure above. The code implementation is as follows: # one pre, one gt def IoU(pred_box, gt_box): ixmin = max(pred_box[0], gt_box[0]) iymin = max(pred_box[1], gt_box[1]) ixmax = min(pred_box[2], gt_box[2]) ...

Added by webspinner on Mon, 03 Jan 2022 07:33:21 +0200

Two discussions on Apriltag processing

Introduction: ※ apriltag detection is a relatively simple method based on visual positioning. This paper discusses the effect of homography matrix in apriltag detection results, which is inconsistent with the application that can be directly used to draw apriltag normal vector. The ability of apriltag algorithm pasted on non plane is tested. If ...

Added by who_cares on Mon, 03 Jan 2022 07:28:53 +0200

Ptorch cifar10 image classification LeNet5

Ptorch cifar10 image classification LeNet5 Here is a summary: Summary 4. Define network (LeNet5) Handwritten font recognition model LeNet5 was born in 1994 and is one of the earliest convolutional neural networks. Through ingenious design, LeNet5 uses convolution, parameter sharing, pooling and other operations to extract features, avoid ...

Added by bongbong on Sun, 02 Jan 2022 23:59:21 +0200

*** Part 2: image preprocessing interface cv::dnn::blobFromImage()

preface When deploying CV algorithm based on deep learning, almost all of them involve input image preprocessing. In python, the task of image preprocessing is basically contracted by numpy. But, when the deployment needs to be implemented in C + +, the problem arises: numpy cannot be used. How to convert the image into the format accept ...

Added by jovanross on Sun, 02 Jan 2022 23:00:37 +0200

How to invoke multiple models in Deepstream

Deepstream provides nvinfer plugin s for encapsulating call models. Deepstream's sample code, deepstream app, implements the use of one primary inference engine for invoking multiple models, and one or more secondary inference engines, because the models are converted to TensorRT's engine files before they are used. So it's generally called eng ...

Added by Francky683 on Sun, 02 Jan 2022 04:11:27 +0200

Triton: openai open source GPU Programming artifact

preface We will release Triton 1.0, an open source Python like programming language that enables researchers without CUDA experience to write efficient GPU code - in most cases equivalent to the code generated by experts. Triton makes it possible to reach the peak of hardware performance with relatively little effort; For example, it can b ...

Added by callie212 on Fri, 31 Dec 2021 20:40:08 +0200

HyperLPR source code analysis 12

2021SC@SDUSC In this analysis, I will briefly introduce and summarize the SimpleRecognizePlate function. Since this function is mainly analyzed by other members of my team, there is no need to analyze it in detail here. The focus is to compare it with the SimpleRecognizePlateByE2E function I analyzed. The function code is as follows: def Sim ...

Added by cassius on Wed, 29 Dec 2021 22:25:19 +0200

Overview of the overall structure of paddedetection

2021SC@SDUSC This article is an overview of the overall structure of paddedetection Structure diagram: The core idea is to form a complete pipline through yaml file by combining the main module and pluggable module for train, eval, infer and export_model. yaml file is equivalent to design drawings. The program will design the structure we ...

Added by JohnM_2000 on Wed, 29 Dec 2021 15:50:10 +0200

Opencv learning 3 Shape recognition module

Shape Detection Contour detection contours, hierarchy = cv2.findContours(image,mode,method) The first parameter input image,The second parameter represents the retrieval mode of contour, which has four types: 1.cv2.RETR_EXTERNAL means that only the outer contour is detected 2.cv2. RETR_ The contour detected by list does not establish a hiera ...

Added by erikjan on Wed, 29 Dec 2021 11:29:23 +0200

3 image coding technology and standard for Python image processing - wavelet transform coding

6.3 image coding technology and standard of Python image processing - wavelet transform coding 1 algorithm principle The so-called wavelet is small for Fourier waves, Fourier wave refers to sine wave (or cosine wave) that oscillates infinitely in time domain. Relatively speaking, wavelet refers to a wave with very concentrated energy in t ...

Added by txmedic03 on Wed, 29 Dec 2021 08:25:53 +0200