title
YOLO Object Detection Using OpenCV And Python | Python Projects | Python Training | Edureka

description
🔥Edureka Python Training: https://www.edureka.co/python-programming-certification-training/ This Edureka video on " YOLO Object Detection Using OpenCV and Python", will walk you through the basics of OpenCV and the implementation of the YOLO algorithm for object detection. The following are covered in this Python & Open CV Tutorial video: 00:00:00 Introduction 00:55:00 Introduction to Computer Vision 00:02:43 A different approach for Object Detection 00:04:27 What is Open-CV? 00:16:25 CNN for Image Processing 00:22:27 YOLO Algorithm 00:24:54 Real-Time object detection using YOLO and Open-CV 🔹Checkout Edureka's Python Tutorial Playlist: https://goo.gl/WsBpKe 🔹Checkout Edureka's Python Tutorial Blog Series: http://bit.ly/2sqmP4s 🔴Do subscribe to our channel and hit the bell icon to never miss an update from us in the future: https://goo.gl/6ohpTV Edureka Community: https://bit.ly/EdurekaCommunity Instagram: https://www.instagram.com/edureka_learning/ Facebook: https://www.facebook.com/edurekaIN/ Twitter: https://twitter.com/edurekain LinkedIn: https://www.linkedin.com/company/edureka Telegram: https://t.me/edurekaupdates SlideShare: https://www.slideshare.net/EdurekaIN Meetup: https://www.meetup.com/edureka/ #Edureka #EdurekaPython #YOLOObjectDetection #PythonProjects #PythonProgramming #OpenCVProjects #PythonTutorial #PythonTraining --------------------------- How it Works? 1. This is a 5 Week Instructor-led Online Course,40 hours of assignment and 20 hours of project work 2. We have a 24x7 One-on-One LIVE Technical Support to help you with any problems you might face or any clarifications you may require during the course. 3. At the end of the training, you will be working on a real-time project for which we will provide you a Grade and a Verifiable Certificate! - - - - - - - - - - - - - - - - - About the Course Edureka's Python Online Certification Training will make you an expert in Python programming. It will also help you learn Python the Big data way with integration of Machine learning, Pig, Hive, and Web Scraping through beautiful soup. During our Python Certification training, our instructors will help you: 1. Master the Basic and Advanced Concepts of Python 2. Understand Python Scripts on UNIX/Windows, Python Editors, and IDEs 3. Master the Concepts of Sequences and File operations 4. Learn how to use and create functions, sorting different elements, Lambda function, error handling techniques, and Regular expressions and using modules in Python 5. Gain expertise in machine learning using Python and build a Real-Life Machine Learning application 6. Understand the supervised and unsupervised learning and concepts of Scikit-Learn 7. Master the concepts of MapReduce in Hadoop 8. Learn to write Complex MapReduce programs 9. Understand what is PIG and HIVE, Streaming feature in Hadoop, MapReduce job running with Python 10. Implementing a PIG UDF in Python, Writing a HIVE UDF in Python, Pydoop, and/Or MRjob Basics 11. Master the concepts of Web scraping in Python 12. Work on a Real Life Project on Big Data Analytics using Python and gain Hands-on Project Experience - - - - - - - - - - - - - - - - - - - Why learn Python? Programmers love Python because of how fast and easy it is to use. Python cuts development time in half with its simple to read syntax and easy compilation feature. Debugging your programs is a breeze in Python with its built-in debugger. Using Python makes Programmers more productive and their programs ultimately better. Python continues to be a favorite option for data scientists who use it for building and using Machine learning applications and other scientific computations. Python runs on Windows, Linux/Unix, Mac OS and has been ported to Java and .NET virtual machines. Python is free to use, even for commercial products, because of its OSI-approved open source license. Python has evolved as the most preferred Language for Data Analytics and the increasing search trends on python also indicate that Python is the next "Big Thing" and a must for Professionals in the Data Analytics domain. Who should go for python? The Python Programming Certification Course is a good fit for the below professionals: Programmers, Developers, Technical Leads, Architects, Freshers Data Scientists, Data Analysts Statisticians and Analysts Business Analysts Project Managers Business Intelligence Managers For more information, please write back to us at sales@edureka.in or call us at IND: 9606058406 / US: 18338555775

detail
{'title': 'YOLO Object Detection Using OpenCV And Python | Python Projects | Python Training | Edureka', 'heatmap': [{'end': 1592.415, 'start': 1533.331, 'weight': 0.762}], 'summary': 'Covers object detection using yolo algorithm and opencv, exploring deep learning techniques, cnn architecture, object detection techniques, and yolo algorithm evolution. it demonstrates yolo algorithm implementation for detecting various objects and adding bounding boxes using opencv, achieving an 89% probability detection on a person and accepting webcam feed, images, and video as inputs.', 'chapters': [{'end': 246.55, 'segs': [{'end': 52.058, 'src': 'embed', 'start': 25.022, 'weight': 0, 'content': [{'end': 29.064, 'text': 'Then we shall discuss both machine learning and deep learning approach for object detection.', 'start': 25.022, 'duration': 4.042}, {'end': 34.167, 'text': "Moving ahead, I'll be speaking about how OpenCV can be used to pre-process our images.", 'start': 29.805, 'duration': 4.362}, {'end': 40.331, 'text': "Finally, we'll discuss about both YOLO and CNN algorithms to object detection and see how we can implement it.", 'start': 34.668, 'duration': 5.663}, {'end': 46.235, 'text': 'Before we begin, consider subscribing to our channel and hit the bell icon to stay updated on trending technologies.', 'start': 40.931, 'duration': 5.304}, {'end': 52.058, 'text': "And also if you're looking for online training certification in Python, check out the link given in the description box below.", 'start': 46.915, 'duration': 5.143}], 'summary': 'Discussion on machine learning, deep learning, opencv for object detection, and promotion of online training in python.', 'duration': 27.036, 'max_score': 25.022, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE25022.jpg'}, {'end': 141.989, 'src': 'embed', 'start': 117.832, 'weight': 2, 'content': [{'end': 124.237, 'text': 'We can perform various tasks like object detection object classification image captioning and image reconstruction.', 'start': 117.832, 'duration': 6.405}, {'end': 127.4, 'text': "I'm sure you might be wondering what does this task do right?", 'start': 124.818, 'duration': 2.582}, {'end': 129.181, 'text': "So let's now discuss each of them.", 'start': 127.74, 'duration': 1.441}, {'end': 135.707, 'text': 'in brief, object detection is ability to detect object or identify object in any given image correctly.', 'start': 129.181, 'duration': 6.526}, {'end': 141.989, 'text': 'Then we have image classification, which basically means to identify what class the object belongs to.', 'start': 136.287, 'duration': 5.702}], 'summary': 'Ai can perform tasks like object detection, classification, and image captioning.', 'duration': 24.157, 'max_score': 117.832, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE117832.jpg'}, {'end': 228.606, 'src': 'embed', 'start': 205.931, 'weight': 3, 'content': [{'end': 214.157, 'text': 'machine learning methods for object detection are sift, then we have support vector machine and then we have Wyler Jones object detection framework.', 'start': 205.931, 'duration': 8.226}, {'end': 217.439, 'text': 'Moving on to the next method that is deep learning.', 'start': 214.978, 'duration': 2.461}, {'end': 222.922, 'text': 'Deep learning which is also referred to as deep structured learning is a class of machine learning algorithm.', 'start': 217.96, 'duration': 4.962}, {'end': 228.606, 'text': 'Deep learning uses multiple layer approach to extract high-level feature from the data that is provided to it.', 'start': 223.423, 'duration': 5.183}], 'summary': 'Object detection methods: sift, svm, wj, and deep learning with multiple layers.', 'duration': 22.675, 'max_score': 205.931, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE205931.jpg'}], 'start': 7.534, 'title': 'Object detection and computer vision basics', 'summary': 'Discusses object detection using yolo algorithm and opencv, covering computer vision, machine learning, and deep learning approaches, as well as the introduction of computer vision tasks and methods used in object detection.', 'chapters': [{'end': 40.331, 'start': 7.534, 'title': 'Object detection with yolo algorithm', 'summary': 'Discusses object detection using yolo algorithm and opencv, covering the agenda of computer vision, machine learning and deep learning approaches, image pre-processing with opencv, and implementation of yolo and cnn algorithms for object detection.', 'duration': 32.797, 'highlights': ['The chapter covers the agenda of computer vision, machine learning, and deep learning approaches for object detection, providing a comprehensive understanding of the topic.', 'The session discusses the use of OpenCV for pre-processing images, offering practical insights into image manipulation and enhancement.', 'It includes a detailed explanation of YOLO and CNN algorithms for object detection, with practical implementation guidance to understand and apply the concepts in real-world scenarios.', 'Junaid from Edureka introduces the session on object detection using YOLO algorithm and OpenCV, setting the context and creating anticipation for the upcoming content.']}, {'end': 246.55, 'start': 40.931, 'title': 'Computer vision basics', 'summary': 'Introduces computer vision, explaining its tasks like object detection, classification, captioning, and reconstruction, and discusses the machine learning and deep learning approaches in object detection, with an overview of the methods used.', 'duration': 205.619, 'highlights': ['The tasks of computer vision include object detection, classification, image captioning, and image reconstruction, each serving a specific purpose in analyzing and understanding images.', 'Machine learning approach in object detection involves the application of algorithms like sift, support vector machine, and Wyler Jones object detection framework, enabling the computer to learn and make predictions based on the given data.', 'Deep learning, a class of machine learning algorithm, uses a multi-layer approach to extract high-level features from the provided data and is influenced by artificial neural networks present in the human brain.']}], 'duration': 239.016, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE7534.jpg', 'highlights': ['The chapter covers the agenda of computer vision, machine learning, and deep learning approaches for object detection, providing a comprehensive understanding of the topic.', 'It includes a detailed explanation of YOLO and CNN algorithms for object detection, with practical implementation guidance to understand and apply the concepts in real-world scenarios.', 'The tasks of computer vision include object detection, classification, image captioning, and image reconstruction, each serving a specific purpose in analyzing and understanding images.', 'Machine learning approach in object detection involves the application of algorithms like sift, support vector machine, and Wyler Jones object detection framework, enabling the computer to learn and make predictions based on the given data.']}, {'end': 978.826, 'segs': [{'end': 288.829, 'src': 'embed', 'start': 247.61, 'weight': 0, 'content': [{'end': 251.011, 'text': 'Most of the deep learning method implement neural network to achieve the results.', 'start': 247.61, 'duration': 3.401}, {'end': 257.851, 'text': 'All the deep learning model require a huge amount of computation power and large volume of label data to learn from the features.', 'start': 251.571, 'duration': 6.28}, {'end': 266.014, 'text': 'Some of the deep learning method for object detection are R-CNN, Faster R-CNN, YOLO algorithm and the Faster R-CNN.', 'start': 258.632, 'duration': 7.382}, {'end': 270.456, 'text': "Moving on, let's speak about open source tool called as OpenCV.", 'start': 266.954, 'duration': 3.502}, {'end': 279.643, 'text': 'OpenCV is a huge open source library for computer vision, machine learning and image processing, and now it plays a major role in real-time operation,', 'start': 271.277, 'duration': 8.366}, {'end': 281.404, 'text': "which is very important in today's system.", 'start': 279.643, 'duration': 1.761}, {'end': 288.829, 'text': 'By using OpenCV, we can pre-process images, videos to identify objects, faces or even handwriting of a human being.', 'start': 281.964, 'duration': 6.865}], 'summary': 'Deep learning methods require large computation power and label data. opencv plays a major role in real-time image processing and object detection.', 'duration': 41.219, 'max_score': 247.61, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE247610.jpg'}, {'end': 512.994, 'src': 'embed', 'start': 484.014, 'weight': 2, 'content': [{'end': 488.921, 'text': 'So let me give it as image, or let me give it as new image, right.', 'start': 484.014, 'duration': 4.907}, {'end': 496.784, 'text': "so it's going to be new image is equal to cv2 dot cvt color.", 'start': 488.921, 'duration': 7.863}, {'end': 503.926, 'text': "okay, and this method takes an image over here and then we're supposed to pass which format we want to convert this to.", 'start': 496.784, 'duration': 7.142}, {'end': 506.527, 'text': "so it's going to be cv2 dot color.", 'start': 503.926, 'duration': 2.601}, {'end': 512.994, 'text': 'Okay, and now we have to convert this from BGR to RGB, so you can see over here.', 'start': 507.611, 'duration': 5.383}], 'summary': 'Convert image from bgr to rgb using cv2.cvtcolor method.', 'duration': 28.98, 'max_score': 484.014, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE484014.jpg'}, {'end': 723.344, 'src': 'embed', 'start': 692.525, 'weight': 4, 'content': [{'end': 694.346, 'text': 'So this is all about splitting our image.', 'start': 692.525, 'duration': 1.821}, {'end': 698.347, 'text': "Now what to do if you want to merge our image, right? So it's pretty simple.", 'start': 694.706, 'duration': 3.641}, {'end': 700.908, 'text': 'We have cv2 dot merge for that.', 'start': 698.407, 'duration': 2.501}, {'end': 703.429, 'text': "So let's give the same name new image.", 'start': 701.408, 'duration': 2.021}, {'end': 707.451, 'text': 'This is going to be cv2 dot merge.', 'start': 704.75, 'duration': 2.701}, {'end': 712.118, 'text': "and we're going to pass our channels and this will be in a form of a tuple.", 'start': 708.556, 'duration': 3.562}, {'end': 715.34, 'text': "So it's going to be R G and B.", 'start': 712.198, 'duration': 3.142}, {'end': 715.64, 'text': 'All right.', 'start': 715.34, 'duration': 0.3}, {'end': 717.581, 'text': 'So this is how we can merge our image.', 'start': 715.98, 'duration': 1.601}, {'end': 723.344, 'text': "Let's see how we can resize the image, right? So let's see another operation wherein we are going to resize the image.", 'start': 718.021, 'duration': 5.323}], 'summary': 'The transcript discusses splitting and merging images using cv2, as well as resizing images.', 'duration': 30.819, 'max_score': 692.525, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE692525.jpg'}, {'end': 863.085, 'src': 'embed', 'start': 826.457, 'weight': 3, 'content': [{'end': 833.241, 'text': 'So, as you can see here, we have reduced our image size, but the number of channels has not changed, but the value,', 'start': 826.457, 'duration': 6.784}, {'end': 840.865, 'text': 'which was something like 720 and 1080, has been reduced to 72 and 182, right?, All right.', 'start': 833.241, 'duration': 7.624}, {'end': 843.726, 'text': 'So let us now see another operation that we can perform here.', 'start': 840.925, 'duration': 2.801}, {'end': 849.129, 'text': "Okay So let's see one more operation that is nothing but rotate operation.", 'start': 843.746, 'duration': 5.383}, {'end': 863.085, 'text': 'okay. so similarly we obviously need to have the height and weight.', 'start': 859.403, 'duration': 3.682}], 'summary': 'Reduced image size to 72x182, discussed rotate operation.', 'duration': 36.628, 'max_score': 826.457, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE826457.jpg'}], 'start': 247.61, 'title': 'Image processing with deep learning and opencv', 'summary': 'Explores deep learning techniques for object detection and emphasizes the role of opencv in real-time image and video processing. it covers operations like image reconstruction, color conversion, splitting, merging, resizing, and rotation, with demonstrations such as converting an image from bgr to rgb and resizing it from 720x1280 to 72x182.', 'chapters': [{'end': 435.125, 'start': 247.61, 'title': 'Deep learning and opencv for image processing', 'summary': 'Discusses the use of deep learning methods like neural networks for object detection, and emphasizes the importance of opencv in real-time operations for image and video processing, showcasing the process of reading and manipulating images using opencv in python.', 'duration': 187.515, 'highlights': ['The chapter discusses the use of deep learning methods like neural networks for object detection Deep learning methods such as R-CNN, Faster R-CNN, and YOLO algorithm are mentioned for object detection.', 'The importance of OpenCV in real-time operations for image and video processing OpenCV is highlighted as a major tool for real-time image processing, enabling pre-processing of images, videos, and identification of objects, faces, or handwriting.', 'Showcasing the process of reading and manipulating images using OpenCV in Python The transcript details the process of reading images using OpenCV, including the installation process, importing libraries, reading images, and understanding image properties like shape and channels.']}, {'end': 978.826, 'start': 435.145, 'title': 'Image processing operations with opencv', 'summary': 'Covers image processing operations including image reconstruction, color conversion, image splitting, merging, resizing, and rotation using opencv, demonstrating the conversion of an image from bgr to rgb and resizing an image from 720x1280 to 72x182.', 'duration': 543.681, 'highlights': ['The chapter demonstrates the conversion of images from BGR to RGB format using the cv2.cvtColor method, showing the color inversion and the process of plotting the image, resulting in a significant difference in the appearance of the image.', 'It explains the process of splitting images into different channels (R, G, B) and printing their shapes, with an example showing the shape of the new image as 720x1280x3 after splitting.', 'The chapter illustrates the process of merging images using the cv2.merge method, providing a simple demonstration of how to merge channels (R, G, B) to reconstruct the original image.', 'It presents the process of resizing an image by a scaling factor of 10, showcasing the calculation of the new dimensions, the use of cv2.resize method, and the reduction in image size from 720x1080 to 72x182 while maintaining the number of channels.', 'The chapter explains the process of rotating an image by 90 degrees counterclockwise, showing the calculation of the center, obtaining the rotation matrix, and using the cv2.warpAffine method to achieve the rotation.']}], 'duration': 731.216, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE247610.jpg', 'highlights': ['The importance of OpenCV in real-time operations for image and video processing OpenCV is highlighted as a major tool for real-time image processing, enabling pre-processing of images, videos, and identification of objects, faces, or handwriting.', 'The chapter discusses the use of deep learning methods like neural networks for object detection Deep learning methods such as R-CNN, Faster R-CNN, and YOLO algorithm are mentioned for object detection.', 'The chapter demonstrates the conversion of images from BGR to RGB format using the cv2.cvtColor method, showing the color inversion and the process of plotting the image, resulting in a significant difference in the appearance of the image.', 'It presents the process of resizing an image by a scaling factor of 10, showcasing the calculation of the new dimensions, the use of cv2.resize method, and the reduction in image size from 720x1080 to 72x182 while maintaining the number of channels.', 'The chapter illustrates the process of merging images using the cv2.merge method, providing a simple demonstration of how to merge channels (R, G, B) to reconstruct the original image.']}, {'end': 1345.801, 'segs': [{'end': 1025.674, 'src': 'embed', 'start': 980.007, 'weight': 0, 'content': [{'end': 983.569, 'text': 'So as you can see here, we have rotated our image counterclockwise by 90 degrees.', 'start': 980.007, 'duration': 3.562}, {'end': 986.969, 'text': 'So moving ahead.', 'start': 986.229, 'duration': 0.74}, {'end': 989.731, 'text': 'Let us now see how convolution neural network work.', 'start': 987.009, 'duration': 2.722}, {'end': 997.114, 'text': 'Okay So what is convolution neural network CNN or a convolution neural network is a class of deep learning neural network.', 'start': 990.671, 'duration': 6.443}, {'end': 1003.557, 'text': "What I'm trying to say here is think of CNN as a machine learning algorithm that can take in an input image,", 'start': 997.854, 'duration': 5.703}, {'end': 1008.299, 'text': 'assign importance to an object and then to be able to differentiate between one object and the other.', 'start': 1003.557, 'duration': 4.742}, {'end': 1016.726, 'text': 'CNN works by extracting features from the images any CNN consists of following three things an input layer, which is a grayscale image.', 'start': 1009.099, 'duration': 7.627}, {'end': 1025.674, 'text': 'Then we have the output layer, which is the binary or multi-class labels, and then we have hidden layers, which contains convolution layer, ReLU,', 'start': 1017.287, 'duration': 8.387}], 'summary': "Rotated image counterclockwise by 90 degrees, explained cnn's function and components.", 'duration': 45.667, 'max_score': 980.007, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE980007.jpg'}, {'end': 1157.87, 'src': 'embed', 'start': 1129.69, 'weight': 3, 'content': [{'end': 1133.013, 'text': 'It was having single channel after undergoing convolution.', 'start': 1129.69, 'duration': 3.323}, {'end': 1135.935, 'text': 'It will have 35 layers or dimensions.', 'start': 1133.073, 'duration': 2.862}, {'end': 1136.455, 'text': 'I should say.', 'start': 1135.995, 'duration': 0.46}, {'end': 1145.482, 'text': 'So this would be 35 and the size over here will be reduced from 10 to 8.', 'start': 1138.076, 'duration': 7.406}, {'end': 1149.525, 'text': "So if you didn't understand this, right, let me quickly make you understand what's happening over here.", 'start': 1145.482, 'duration': 4.043}, {'end': 1154.128, 'text': "So what's going to happen over here is this filters over here, right? This 3 cross 3 filters.", 'start': 1149.985, 'duration': 4.143}, {'end': 1157.87, 'text': 'They will go on each and every part of this image.', 'start': 1155.506, 'duration': 2.364}], 'summary': 'After convolution, it will have 35 layers with a reduction in size from 10 to 8.', 'duration': 28.18, 'max_score': 1129.69, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1129690.jpg'}, {'end': 1281.301, 'src': 'embed', 'start': 1252.203, 'weight': 4, 'content': [{'end': 1255.024, 'text': 'Similarly, this will go through all each and every box.', 'start': 1252.203, 'duration': 2.821}, {'end': 1260.642, 'text': 'All right, and now the size of the image will reduce.', 'start': 1257.981, 'duration': 2.661}, {'end': 1264.264, 'text': 'Okay, So let me quickly show you by what.', 'start': 1261.343, 'duration': 2.921}, {'end': 1266.805, 'text': 'so now we are going to pass the max pool layer.', 'start': 1264.264, 'duration': 2.541}, {'end': 1269.627, 'text': 'after this, the size of the image will be reduced by half.', 'start': 1266.805, 'duration': 2.822}, {'end': 1273.509, 'text': "that's going to be 4, cross 4, but number of filters will not change.", 'start': 1269.627, 'duration': 3.882}, {'end': 1278.678, 'text': 'All right.', 'start': 1278.137, 'duration': 0.541}, {'end': 1281.301, 'text': 'now we can pass this through another convolution layer.', 'start': 1278.678, 'duration': 2.623}], 'summary': 'Image size reduced by half after max pool layer, remaining 4x4, filters unchanged.', 'duration': 29.098, 'max_score': 1252.203, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1252203.jpg'}, {'end': 1355.306, 'src': 'embed', 'start': 1324.029, 'weight': 5, 'content': [{'end': 1330.052, 'text': 'So what ReLU does is this is how the graph for ReLU does, and here it ranges from 0 and 1.', 'start': 1324.029, 'duration': 6.023}, {'end': 1334.154, 'text': 'and now, in order to perform the classification in our last output layer right?', 'start': 1330.052, 'duration': 4.102}, {'end': 1336.916, 'text': 'I will be passing an activation function called as softmax.', 'start': 1334.274, 'duration': 2.642}, {'end': 1343.259, 'text': 'Because softmax gives me the probability which would range from minus 1 to plus 1.', 'start': 1338.217, 'duration': 5.042}, {'end': 1343.5, 'text': 'All right.', 'start': 1343.259, 'duration': 0.241}, {'end': 1345.801, 'text': 'So this is all about convolution neural network.', 'start': 1343.68, 'duration': 2.121}, {'end': 1355.306, 'text': 'Using deep learning we can detect objects either by using our CNN model which stands for region-based convolution network or by using Yolo method.', 'start': 1347.21, 'duration': 8.096}], 'summary': 'Relu ranges from 0 to 1, softmax provides probability, cnn detects objects using different methods.', 'duration': 31.277, 'max_score': 1324.029, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1324029.jpg'}], 'start': 980.007, 'title': 'Cnn architecture and basics', 'summary': 'Discusses the functionality of cnn, including image rotation, layers, and neural network for classification. it also covers basics like convolution, max pooling, dimensionality, and activation functions with specific examples.', 'chapters': [{'end': 1044.261, 'start': 980.007, 'title': 'Cnn architecture and functionality', 'summary': 'Explains the functionality of convolution neural networks (cnn), including its ability to rotate images counterclockwise by 90 degrees, and its components such as input layer, output layer, hidden layers, and artificial neural network for classification.', 'duration': 64.254, 'highlights': ['CNN works by extracting features from images and consists of input layer, output layer, hidden layers with convolution, ReLU, and pooling layers, and artificial neural network for classification.', 'CNN can rotate images counterclockwise by 90 degrees.', 'CNN is a class of deep learning neural network used for assigning importance to objects in input images and differentiating between objects.']}, {'end': 1345.801, 'start': 1044.461, 'title': 'Convolution neural network basics', 'summary': 'Explains the basics of convolutional neural network (cnn) including concepts like convolution, max pooling, curse of dimensionality, and activation functions, with examples of image size reduction and feature extraction using 3x3 filters and 2x2 max pooling, ultimately leading to classification using relu and softmax functions.', 'duration': 301.34, 'highlights': ['Explaining the process of convolution and dimensionality reduction in CNN using 3x3 filters and 35 layers to extract features. 35 layers, image size reduced from 10x10 to 8x8', 'Describing the concept and purpose of max pooling in CNN to reduce image size by half using 2x2 max pool layer. Image size reduced from 8x8 to 4x4', 'Introducing the use of ReLU activation function in CNN for increasing linearity and explaining its graph and range. ReLU activation function range: 0 to 1', 'Mentioning the application of softmax activation function for classification and its probability range. Softmax activation function probability range: -1 to 1']}], 'duration': 365.794, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE980007.jpg', 'highlights': ['CNN consists of input, output, hidden layers with convolution, ReLU, pooling, and neural network for classification.', 'CNN can rotate images counterclockwise by 90 degrees.', 'CNN assigns importance to objects in input images and differentiates between objects.', '35 layers and 3x3 filters are used in CNN for dimensionality reduction and feature extraction.', 'Max pooling in CNN reduces image size by half using 2x2 max pool layer.', 'ReLU activation function in CNN increases linearity with a range of 0 to 1.', 'Softmax activation function in CNN is used for classification with a probability range of -1 to 1.']}, {'end': 1617.497, 'segs': [{'end': 1400.898, 'src': 'embed', 'start': 1363.012, 'weight': 3, 'content': [{'end': 1369.82, 'text': "right?. Well, you see, there's a lot of difference between Yolo family and RCNN based approach in the RCNN based approach.", 'start': 1363.012, 'duration': 6.808}, {'end': 1377.789, 'text': 'It focuses mostly on division of an image into parts and then assign probability values to those part and whichever part has a highest probability.', 'start': 1369.88, 'duration': 7.909}, {'end': 1380.031, 'text': "It's where we consider an object to be present.", 'start': 1378.129, 'duration': 1.902}, {'end': 1389.214, 'text': 'Whereas the Yolo framework focuses on the entire image as a whole and predicts the bounding boxes and then calculate the class probability to label the boxes.', 'start': 1380.652, 'duration': 8.562}, {'end': 1395.896, 'text': 'The family of Yolo framework is very fast as compared to our CNN Yolo algorithm has evolved over the years.', 'start': 1389.675, 'duration': 6.221}, {'end': 1397.537, 'text': 'It first started with Yolo v1.', 'start': 1396.276, 'duration': 1.261}, {'end': 1400.898, 'text': 'This model is also called as Yolo unified,', 'start': 1398.397, 'duration': 2.501}], 'summary': 'Rcnn focuses on image parts, yolo predicts bounding boxes for entire image, yolo framework is faster and has evolved over the years.', 'duration': 37.886, 'max_score': 1363.012, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1363012.jpg'}, {'end': 1487.676, 'src': 'embed', 'start': 1419.348, 'weight': 0, 'content': [{'end': 1423.171, 'text': 'Now we have YOLOv2 and the latest version of YOLO is YOLOv3.', 'start': 1419.348, 'duration': 3.823}, {'end': 1431.599, 'text': 'You see the YOLOv1 framework makes several localization error and YOLOv2 improves this by focusing on recall and localization.', 'start': 1423.872, 'duration': 7.727}, {'end': 1441.085, 'text': 'The YOLOv2 uses batch normalization, anchored boxes, high resolution classifiers, fine gradient features and multi-level classification,', 'start': 1432.299, 'duration': 8.786}, {'end': 1443.186, 'text': 'and also it uses something called as Darknet.', 'start': 1441.085, 'duration': 2.101}, {'end': 1446.509, 'text': 'All these features made YOLOv2 better than v1.', 'start': 1443.787, 'duration': 2.722}, {'end': 1448.27, 'text': 'Speaking about Darknet.', 'start': 1446.969, 'duration': 1.301}, {'end': 1455.575, 'text': 'Darknet is actually a pre-trained model and here YOLOv2 was using Darknet 19,, which means it contains 19 convolution layer,', 'start': 1448.27, 'duration': 7.305}, {'end': 1459.317, 'text': '5 maxpool layer and a softmax layer for object classification.', 'start': 1455.575, 'duration': 3.742}, {'end': 1461.679, 'text': 'The latest model of YOLO is YOLOv3.', 'start': 1459.897, 'duration': 1.782}, {'end': 1466.022, 'text': 'This model is a fastest and most accurate object detection model.', 'start': 1462.619, 'duration': 3.403}, {'end': 1472.686, 'text': 'It accurately classifies the object by using logistic classification compared to soft Mac, which was used in Yolo V2.', 'start': 1466.462, 'duration': 6.224}, {'end': 1476.809, 'text': 'This makes us capable of making multi-label classification.', 'start': 1473.327, 'duration': 3.482}, {'end': 1481.252, 'text': 'Yolo V3, which is also uses Darknet 55 as a feature extractor.', 'start': 1476.809, 'duration': 4.443}, {'end': 1482.033, 'text': 'you see over here.', 'start': 1481.252, 'duration': 0.781}, {'end': 1487.676, 'text': 'Yolo V3 makes use of Darknet 53, which means that there are 53 convolution layer.', 'start': 1482.033, 'duration': 5.643}], 'summary': 'Yolov2 improves on yolov1 with features like batch normalization and anchored boxes, while yolov3 is the fastest and most accurate object detection model.', 'duration': 68.328, 'max_score': 1419.348, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1419348.jpg'}, {'end': 1592.415, 'src': 'heatmap', 'start': 1533.331, 'weight': 0.762, 'content': [{'end': 1536.732, 'text': 'Okay, and this is the official website for our Yolo algorithm.', 'start': 1533.331, 'duration': 3.401}, {'end': 1539.974, 'text': 'And as you can see here, we have multiple versions.', 'start': 1537.573, 'duration': 2.401}, {'end': 1546.976, 'text': 'We have different, different versions over here and then we have Yolo v3 and then we have various versions right?', 'start': 1540.834, 'duration': 6.142}, {'end': 1552.839, 'text': 'So, in order to use this Yolo algorithm, we need to download this weights folder and also the configuration file.', 'start': 1547.216, 'duration': 5.623}, {'end': 1556.08, 'text': 'The configuration file is basically a GitHub repository.', 'start': 1553.399, 'duration': 2.681}, {'end': 1562.787, 'text': "So let me quickly show you that as I've already installed this I won't be reinstalling this on my system, but it's pretty similar.", 'start': 1556.52, 'duration': 6.267}, {'end': 1567.412, 'text': 'All you need to do is click on weights and it automatically download this on your folder.', 'start': 1562.867, 'duration': 4.545}, {'end': 1572.217, 'text': 'And as you can see, I have already installed this weights folder as well as a configuration file.', 'start': 1568.233, 'duration': 3.984}, {'end': 1577.263, 'text': 'All right, so let me upload these files from my folder.', 'start': 1574.18, 'duration': 3.083}, {'end': 1582.849, 'text': 'Okay, So, as you can see, here we have the Cocoa names.', 'start': 1580.327, 'duration': 2.522}, {'end': 1585.811, 'text': 'Cocoa names is nothing but the classes that have been pre-trained right?', 'start': 1582.849, 'duration': 2.962}, {'end': 1587.432, 'text': "So we'll have a Cocoa names.", 'start': 1585.831, 'duration': 1.601}, {'end': 1592.415, 'text': "then we'll have the configuration file as well as the weights, and let me quickly open this.", 'start': 1587.432, 'duration': 4.983}], 'summary': 'Official website for yolo algorithm with multiple versions and downloadable weights and configuration files.', 'duration': 59.084, 'max_score': 1533.331, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1533331.jpg'}], 'start': 1347.21, 'title': 'Object detection techniques and evolution of yolo algorithm', 'summary': "Discusses the differences between rcnn and yolo algorithm for object detection, emphasizing rcnn's image division and probability assignment. it also covers the evolution of yolo algorithm from yolov1 to yolov3, highlighting improvements in processing speed, localization accuracy, and darknet utilization.", 'chapters': [{'end': 1380.031, 'start': 1347.21, 'title': 'Object detection techniques', 'summary': 'Explains the differences between region-based convolution network (rcnn) and yolo algorithm for object detection, highlighting the focus of rcnn on dividing images into parts and assigning probability values.', 'duration': 32.821, 'highlights': ['Yolo algorithm focuses on the entirety of an image, providing a more unified approach to object detection and outperforming RCNN in terms of speed and efficiency.', 'RCNN method involves dividing an image into parts and assigning probability values, while considering the part with the highest probability as the presence of an object.']}, {'end': 1617.497, 'start': 1380.652, 'title': 'Evolution of yolo algorithm', 'summary': 'Explains the evolution of the yolo algorithm, from yolov1 to yolov3, with improvements such as faster processing, fewer localization errors, and the use of darknet for feature extraction.', 'duration': 236.845, 'highlights': ['YOLOv3 is the fastest and most accurate object detection model compared to YOLOv1 and YOLOv2. YOLOv3 is the latest and fastest object detection model which accurately classifies objects using logistic classification, making it capable of multi-label classification.', 'YOLOv2 improves localization errors by focusing on recall and localization, using batch normalization, anchored boxes, high resolution classifiers, fine gradient features, and multi-level classification. YOLOv2 improves localization errors by focusing on recall and localization, using batch normalization, anchored boxes, high resolution classifiers, fine gradient features, and multi-level classification.', 'Darknet 53 in YOLOv3 allows for more accurate predictions of objects due to the 53 convolution layers. YOLOv3 uses Darknet 53 as a feature extractor, which enables more accurate predictions of objects with 53 convolution layers.']}], 'duration': 270.287, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1347210.jpg', 'highlights': ['YOLOv3 is the fastest and most accurate object detection model compared to YOLOv1 and YOLOv2.', 'YOLOv3 uses Darknet 53 as a feature extractor, enabling more accurate predictions of objects with 53 convolution layers.', 'YOLOv2 improves localization errors by focusing on recall and localization, using batch normalization, anchored boxes, high resolution classifiers, fine gradient features, and multi-level classification.', 'RCNN method involves dividing an image into parts and assigning probability values, while considering the part with the highest probability as the presence of an object.', 'Yolo algorithm focuses on the entirety of an image, providing a more unified approach to object detection and outperforming RCNN in terms of speed and efficiency.']}, {'end': 2285.329, 'segs': [{'end': 1646.506, 'src': 'embed', 'start': 1617.957, 'weight': 2, 'content': [{'end': 1620.58, 'text': "So let's give the name of variable over here as net.", 'start': 1617.957, 'duration': 2.623}, {'end': 1624.363, 'text': "Okay, let's give it as Network or we can give it here as Yolo.", 'start': 1621.78, 'duration': 2.583}, {'end': 1633.054, 'text': 'Okay, and this would be CV to dot DNN which stands for deep neural network dot read Network.', 'start': 1625.223, 'duration': 7.831}, {'end': 1636.898, 'text': 'All right, and now here we are going to pass an argument over here.', 'start': 1634.155, 'duration': 2.743}, {'end': 1640.06, 'text': 'We are going to pass our weights as well as a configuration.', 'start': 1636.978, 'duration': 3.082}, {'end': 1646.506, 'text': 'All right, so let me quickly copy the path for this and paste it over here like this.', 'start': 1640.621, 'duration': 5.885}], 'summary': 'Setting up a deep neural network named yolo with weights and configuration.', 'duration': 28.549, 'max_score': 1617.957, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1617957.jpg'}, {'end': 1716.085, 'src': 'embed', 'start': 1683.496, 'weight': 0, 'content': [{'end': 1689.84, 'text': 'right?. It has the list of names that are your algorithm is capable of detecting and, as is a pre-trained model,', 'start': 1683.496, 'duration': 6.344}, {'end': 1691.741, 'text': 'There are a couple of things we can detect cats.', 'start': 1689.98, 'duration': 1.761}, {'end': 1695.304, 'text': 'We can detect dogs horse sheep cow elephant bear and many more.', 'start': 1691.781, 'duration': 3.523}, {'end': 1698.466, 'text': 'Let me know quickly import that and put them in the form of a list.', 'start': 1695.884, 'duration': 2.582}, {'end': 1701.188, 'text': 'So I have already imported matplotlib over here.', 'start': 1699.146, 'duration': 2.042}, {'end': 1703.589, 'text': "So let's create a class CLA.", 'start': 1701.548, 'duration': 2.041}, {'end': 1706.091, 'text': 'And this would be an empty list.', 'start': 1705.03, 'duration': 1.061}, {'end': 1716.085, 'text': "All right, and now what we'll do is as it's a file will use this file handling with open and we'll give the name of this place or that is a path.", 'start': 1707.079, 'duration': 9.006}], 'summary': 'Algorithm capable of detecting various animals, using a pre-trained model to import and list them.', 'duration': 32.589, 'max_score': 1683.496, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1683496.jpg'}, {'end': 1783.388, 'src': 'embed', 'start': 1754.895, 'weight': 3, 'content': [{'end': 1759.078, 'text': 'So it should approximately show around 80 and if you want to see what this classes contain.', 'start': 1754.895, 'duration': 4.183}, {'end': 1764.142, 'text': "They'll contain the name of the different different objects that we can identify.", 'start': 1760.82, 'duration': 3.322}, {'end': 1765.443, 'text': 'All right.', 'start': 1765.183, 'duration': 0.26}, {'end': 1769.704, 'text': 'So let me change this back to the length and execute this.', 'start': 1766.363, 'duration': 3.341}, {'end': 1775.326, 'text': 'So now that we have a classes and we have loaded our model over here.', 'start': 1771.444, 'duration': 3.882}, {'end': 1777.706, 'text': "So what we'll do is let's load our image.", 'start': 1775.726, 'duration': 1.98}, {'end': 1779.487, 'text': "So let's take the same old image.", 'start': 1778.086, 'duration': 1.401}, {'end': 1783.388, 'text': "Let's take this bus image over here and in order to load our image.", 'start': 1779.787, 'duration': 3.601}], 'summary': 'Transcript covers loading classes, model, and image for object identification.', 'duration': 28.493, 'max_score': 1754.895, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1754895.jpg'}, {'end': 1904.968, 'src': 'embed', 'start': 1880.459, 'weight': 1, 'content': [{'end': 1887.807, 'text': "so it's going to be 320, cross 320, all right, and then we'll have to pass the images 0, comma 0,", 'start': 1880.459, 'duration': 7.348}, {'end': 1898.504, 'text': 'comma 0 and then an argument over here which would say swap is equal to true And the reason why we are using the swap RB.', 'start': 1887.807, 'duration': 10.697}, {'end': 1904.968, 'text': "The reason is because we all know it's this is going to read in BGR right and we obviously have to interchange RG and B.", 'start': 1898.624, 'duration': 6.344}], 'summary': 'Using swap argument to interchange rg and b in the images.', 'duration': 24.509, 'max_score': 1880.459, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1880459.jpg'}], 'start': 1617.957, 'title': 'Implementing yolo algorithm and file handling for image processing', 'summary': 'Demonstrates yolo algorithm implementation using opencv to detect various objects, such as cats, dogs, horses, sheep, cows, elephants, and bears from an uploaded weights folder and configuration file. it also covers file handling, image loading, and processing using cv2 library, including operations such as resizing, converting bgr to rgb format, setting input and output layers, and extracting bounding box coordinates with confidence thresholds for object detection in images.', 'chapters': [{'end': 1703.589, 'start': 1617.957, 'title': 'Implementing yolo algorithm with opencv', 'summary': 'Demonstrates the implementation of yolo algorithm using opencv to detect various objects including cats, dogs, horses, sheep, cows, elephants, and bears from an uploaded weights folder and configuration file.', 'duration': 85.632, 'highlights': ['The chapter demonstrates the implementation of YOLO algorithm using OpenCV to detect various objects including cats, dogs, horses, sheep, cows, elephants, and bears from an uploaded weights folder and configuration file.', "The variable 'net' is defined to represent the YOLO neural network, and the uploaded weights and configuration file are passed as arguments to the network.", "The chapter also highlights the import of classes from 'Coco', which contains the list of names that the algorithm can detect, including cats, dogs, horses, sheep, cows, elephants, and bears."]}, {'end': 2285.329, 'start': 1705.03, 'title': 'File handling and image processing', 'summary': 'Covers file handling using open and read operations, image loading and processing using cv2 library, including resizing, converting bgr to rgb format, setting input and output layers, and extracting bounding box coordinates with confidence thresholds for object detection in images.', 'duration': 580.299, 'highlights': ['Using file handling with open, defining read operation and splitting text into lines resulted in approximately 80 classes being identified.', 'Loading and processing an image involved reading in BGR format, converting to RGB format, and reshaping the image to 320x320 size, ensuring values ranging from 0 to 1 and using swapRB parameter to interchange RG and B channels.', 'Setting input image using Yolo model, defining output layer names, and passing the image for forward propagation.', 'Extracting bounding box coordinates, confidence scores, and class IDs for object detection, setting a confidence threshold of 0.7 to prevent multiple bounding boxes, and calculating the center coordinates and dimensions of the bounding boxes for object localization.']}], 'duration': 667.372, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE1617957.jpg', 'highlights': ['Demonstrates YOLO algorithm implementation using OpenCV to detect various objects including cats, dogs, horses, sheep, cows, elephants, and bears from an uploaded weights folder and configuration file.', 'Loading and processing an image involves reading in BGR format, converting to RGB format, and reshaping the image to 320x320 size, ensuring values ranging from 0 to 1 and using swapRB parameter to interchange RG and B channels.', "The variable 'net' is defined to represent the YOLO neural network, and the uploaded weights and configuration file are passed as arguments to the network.", 'Using file handling with open, defining read operation and splitting text into lines resulted in approximately 80 classes being identified.']}, {'end': 2828.39, 'segs': [{'end': 2313.978, 'src': 'embed', 'start': 2285.749, 'weight': 0, 'content': [{'end': 2289.75, 'text': 'And now finally, all we need to do is we need to append these values to these boxes over here.', 'start': 2285.749, 'duration': 4.001}, {'end': 2295.072, 'text': "So what we'll do is boxes that is nothing but these layer.", 'start': 2290.411, 'duration': 4.661}, {'end': 2297.673, 'text': 'So it will be Bo X should be.', 'start': 2295.372, 'duration': 2.301}, {'end': 2301.714, 'text': "Yes, that shouldn't be an issue dot append.", 'start': 2297.713, 'duration': 4.001}, {'end': 2308.716, 'text': "We'll pass a tuple value here of X Y width and height.", 'start': 2303.415, 'duration': 5.301}, {'end': 2313.978, 'text': 'Similarly, we are going to pass the confidence values that is nothing but this tuple confidence.', 'start': 2309.336, 'duration': 4.642}], 'summary': 'Append values to boxes and layers, passing tuple of x y width and height, and confidence values.', 'duration': 28.229, 'max_score': 2285.749, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE2285749.jpg'}, {'end': 2388.692, 'src': 'embed', 'start': 2356.963, 'weight': 4, 'content': [{'end': 2361.545, 'text': "What we'll do is we'll find the list or we'll just print the amount of elements that are present in the list.", 'start': 2356.963, 'duration': 4.582}, {'end': 2364.706, 'text': "Okay So what we'll do is length of boxes.", 'start': 2362.025, 'duration': 2.681}, {'end': 2372.463, 'text': 'Okay, and this should give us okay to that means whatever image we are passing it is able to detect only two objects over there.', 'start': 2365.819, 'duration': 6.644}, {'end': 2373.383, 'text': 'All right.', 'start': 2373.083, 'duration': 0.3}, {'end': 2378.786, 'text': "And now what we'll do is we have to obviously add this bonding boxes to our image, right? So we'll do indices.", 'start': 2373.703, 'duration': 5.083}, {'end': 2388.692, 'text': 'And this will be CV to dot deep neural network dot NMS boxes.', 'start': 2381.007, 'duration': 7.685}], 'summary': 'Detects two objects and adds bounding boxes to the image using deep neural network.', 'duration': 31.729, 'max_score': 2356.963, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE2356963.jpg'}, {'end': 2511.366, 'src': 'embed', 'start': 2480.433, 'weight': 5, 'content': [{'end': 2481.494, 'text': 'And let me execute this now.', 'start': 2480.433, 'duration': 1.061}, {'end': 2487.059, 'text': "Perfect Now, what we'll do is we'll take each and every object over there and we'll add bounding boxes to this.", 'start': 2482.318, 'duration': 4.741}, {'end': 2493.841, 'text': "And in order to do that, we'll use a for loop for I in indexes dot flatten.", 'start': 2487.579, 'duration': 6.262}, {'end': 2496.242, 'text': 'All right.', 'start': 2495.982, 'duration': 0.26}, {'end': 2498.822, 'text': 'And now we have to take our XY coordinates.', 'start': 2496.542, 'duration': 2.28}, {'end': 2503.724, 'text': 'So X comma Y comma width and height.', 'start': 2499.703, 'duration': 4.021}, {'end': 2505.944, 'text': 'This would be nothing but boxes.', 'start': 2503.924, 'duration': 2.02}, {'end': 2508.945, 'text': 'And then this is for that particular image.', 'start': 2507.105, 'duration': 1.84}, {'end': 2511.366, 'text': "Right And now let's have a label.", 'start': 2509.225, 'duration': 2.141}], 'summary': 'Using a for loop, bounding boxes are added to objects with xy coordinates for each image.', 'duration': 30.933, 'max_score': 2480.433, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE2480433.jpg'}, {'end': 2577.219, 'src': 'embed', 'start': 2553.46, 'weight': 1, 'content': [{'end': 2561.407, 'text': 'Yep, So what is happening over here is we are just trying to extract each and every information from the list that we have created above,', 'start': 2553.46, 'duration': 7.947}, {'end': 2562.108, 'text': "and now we'll.", 'start': 2561.407, 'duration': 0.701}, {'end': 2564.97, 'text': "all we'll do is we'll add rectangular bounding boxes to do that.", 'start': 2562.108, 'duration': 2.862}, {'end': 2566.652, 'text': 'We have cv2 dot rectangle.', 'start': 2565.01, 'duration': 1.642}, {'end': 2577.219, 'text': "and then we'll have to define the image, the coordinates, that is x, comma, y, and then we obviously have to define the edges right?", 'start': 2568.214, 'duration': 9.005}], 'summary': 'Extracting information from a list and adding rectangular bounding boxes using cv2.rectangle.', 'duration': 23.759, 'max_score': 2553.46, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE2553460.jpg'}, {'end': 2803.885, 'src': 'embed', 'start': 2773.873, 'weight': 2, 'content': [{'end': 2775.655, 'text': "so it's going to detect a lot of people, right?", 'start': 2773.873, 'duration': 1.782}, {'end': 2780.619, 'text': "So that's why it has given a single bounding box and it has detected a person over here.", 'start': 2776.215, 'duration': 4.404}, {'end': 2789.917, 'text': 'All right, as I mentioned earlier using open CV, we can take three types of inputs right one is a webcam feed and the one is the images.', 'start': 2781.892, 'duration': 8.025}, {'end': 2790.917, 'text': 'Another one is a video.', 'start': 2789.997, 'duration': 0.92}, {'end': 2793.499, 'text': 'All those, the procedure remains the same.', 'start': 2791.398, 'duration': 2.101}, {'end': 2798.422, 'text': "only thing that you're going to do is you're going to change how you take the input and put all of this in the form of a for loop.", 'start': 2793.499, 'duration': 4.923}, {'end': 2801.744, 'text': 'All right guys with this we have come to the end of a session.', 'start': 2799.442, 'duration': 2.302}, {'end': 2803.885, 'text': 'I hope you enjoyed and learn something new.', 'start': 2802.204, 'duration': 1.681}], 'summary': 'Opencv can detect people from webcam feed, images, and videos using a single bounding box. the session covered the process and emphasized utilizing for loops for different inputs.', 'duration': 30.012, 'max_score': 2773.873, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE2773873.jpg'}], 'start': 2285.749, 'title': 'Adding bounding boxes and opencv image processing', 'summary': 'Covers appending bounding box coordinates, confidence values, and class ids, implementing non-maximum suppression, and adding bounding boxes to detect two objects. it also discusses adding rectangular bounding boxes and text using opencv, achieving a 89% probability detection on a person, and accepting webcam feed, images, and video as inputs.', 'chapters': [{'end': 2551.058, 'start': 2285.749, 'title': 'Adding bounding boxes to images', 'summary': 'Illustrates the process of appending bounding box coordinates, confidence values, and class ids to a list, then implementing non-maximum suppression to reduce the number of bounding boxes and adding bounding boxes to the image using a for loop, resulting in the detection of two objects with specific coordinates and labels.', 'duration': 265.309, 'highlights': ['The chapter demonstrates appending bounding box coordinates, confidence values, and class IDs to a list, eventually detecting two objects with specific coordinates and labels.', 'It illustrates the implementation of non-maximum suppression to reduce the number of bounding boxes in the image.', 'The process involves adding bounding boxes to the image using a for loop, with specific coordinates and labels for the detected objects.']}, {'end': 2828.39, 'start': 2553.46, 'title': 'Opencv image processing', 'summary': 'Discusses the process of adding rectangular bounding boxes and text to detect objects in images using opencv, achieving a 89% probability detection on a person and highlighting the capability of taking three types of inputs: webcam feed, images, and video.', 'duration': 274.93, 'highlights': ['The process involves adding rectangular bounding boxes to the image using cv2.rectangle with defined coordinates, width, and color, followed by adding text using cv2.putText with label, confidence value, font, and background parameters.', 'The detection process achieved a 89% probability detection on a person, showcasing the effectiveness of the OpenCV implementation.', 'The capability of OpenCV to take three types of inputs: webcam feed, images, and video, with the procedure remaining the same, except for changing how the input is taken and putting it in a for loop.']}], 'duration': 542.641, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/b59xfUZZqJE/pics/b59xfUZZqJE2285749.jpg', 'highlights': ['The chapter demonstrates appending bounding box coordinates, confidence values, and class IDs to a list, eventually detecting two objects with specific coordinates and labels.', 'The process involves adding rectangular bounding boxes to the image using cv2.rectangle with defined coordinates, width, and color, followed by adding text using cv2.putText with label, confidence value, font, and background parameters.', 'The detection process achieved a 89% probability detection on a person, showcasing the effectiveness of the OpenCV implementation.', 'The capability of OpenCV to take three types of inputs: webcam feed, images, and video, with the procedure remaining the same, except for changing how the input is taken and putting it in a for loop.', 'It illustrates the implementation of non-maximum suppression to reduce the number of bounding boxes in the image.', 'The process involves adding bounding boxes to the image using a for loop, with specific coordinates and labels for the detected objects.']}], 'highlights': ['The chapter covers the agenda of computer vision, machine learning, and deep learning approaches for object detection, providing a comprehensive understanding of the topic.', 'The tasks of computer vision include object detection, classification, image captioning, and image reconstruction, each serving a specific purpose in analyzing and understanding images.', 'The importance of OpenCV in real-time operations for image and video processing OpenCV is highlighted as a major tool for real-time image processing, enabling pre-processing of images, videos, and identification of objects, faces, or handwriting.', 'The chapter discusses the use of deep learning methods like neural networks for object detection Deep learning methods such as R-CNN, Faster R-CNN, and YOLO algorithm are mentioned for object detection.', 'YOLOv3 is the fastest and most accurate object detection model compared to YOLOv1 and YOLOv2.', 'Demonstrates YOLO algorithm implementation using OpenCV to detect various objects including cats, dogs, horses, sheep, cows, elephants, and bears from an uploaded weights folder and configuration file.', 'The process involves adding rectangular bounding boxes to the image using cv2.rectangle with defined coordinates, width, and color, followed by adding text using cv2.putText with label, confidence value, font, and background parameters.', 'The detection process achieved a 89% probability detection on a person, showcasing the effectiveness of the OpenCV implementation.']}