Lecture 16: Detection and Segmentation
32.6 هزار بار بازدید -
4 سال پیش
-
Lecture 16 continues our discussion
Lecture 16 continues our discussion of localizing objects in images with neural networks. We recap the R-CNN family of methods from the previous lecture and discuss how these networks behave differently during training and testing. We briefly discuss anchor-free object detectors such as CornerNet. We then move on to other localization tasks, and see how fully-convolutional networks can be used for semantic segmentation. This leads to a discussion of different methods of upsampling in neural networks, including max unpooling and transposed convolution. We then discuss the task of instance segmentation, where neural networks must both identify the object instances in each image as well as the pixels belonging to each object. We see how the Mask R-CNN architecture generalizes Faster R-CNN to perform instance segmentation. We briefly discuss other localization tasks including panoptic segmentation and keypoint estimation.
Slides: http://myumi.ch/2Del3
_________________________________________________________________________________________________
Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks.
Course Website: http://myumi.ch/Bo9Ng
Instructor: Justin Johnson http://myumi.ch/QA8Pg
Slides: http://myumi.ch/2Del3
_________________________________________________________________________________________________
Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks.
Course Website: http://myumi.ch/Bo9Ng
Instructor: Justin Johnson http://myumi.ch/QA8Pg
4 سال پیش
در تاریخ 1399/05/20 منتشر شده
است.
32,618
بـار بازدید شده