Lecture 22: Conclusion

Lecture 22 gives a summary and review of all the topics we covered this semester. After this summary, I spend the second half of the lecture talking about my thoughts on the future of computer vision and deep learning. We predict that new neural network architectures will continue to surprise us, such as Neural ODEs. Neural networks will continue to find new and surprising applications, such as learned data structures and symbolic mathematics. Deep networks will continue to use ever-increasing amounts of data and compute, and will take advantage of specialized hardware. We also identify some important open problems in computer vision including bias, lack of theoretical understanding, the need for low-shot and self-supervised learning algorithms, and the lack of “common sense” understanding in deep networks. Note: This lecture briefly touches on the ideas of bias and fairness in machine learning and computer vision models. This is an important topic that I plan to expand in future offerings of this course. In the meantime I recommend watching the videos from the CVPR 2020 Tutorial on Fairness, Accountability, Transparency, and Ethics in Computer Vision (FATE/CV): https://sites.google.com/view/fatecv-tutorial/schedule Slides: http://myumi.ch/dO1bK _________________________________________________________________________________________________ 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