Lecture 1: Introduction to Deep Learning for Computer Vision
Vložit
- čas přidán 6. 06. 2024
- Lecture 1 gives a broad introduction to computer vision and machine learning. We give a brief history of the two fields, starting in the 1950s and leading up to the modern explosion of deep neural networks. We preview some of the topics we will cover in the rest of the course, and discuss the enormous potential of deep learning and computer vision to improve our lives. We also discuss the logistics and philosophy of this course.
Slides: myumi.ch/yKgM3
_________________________________________________________________________________________________
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: myumi.ch/Bo9Ng
Instructor: Justin Johnson myumi.ch/QA8Pg
This has to one of the best intros to CV ever! I'm in awe of the concise summary of the most influential papers in this field.
Can't believe he was teaching for the first time and he managed to make it so good and interesting. Really good job.
Easily one of the best current online courses for Deep Learning. Not only is this guy clearly an expert but he is just so enthusiastic about the subject. He doesn't just lecture but he actually teaches you. I love it when he says 'more concretely'. That always reminds me of Andrew Ng - another brilliant teacher of machine learning. Andrew's lectures are legendary! Anyway, well done Justin - keep up the good work!
Standing ovation for Michigan Online casually dropping one of the clearest, best explanations of deep learning online, all for free! We truly live in amazing times
Thank you so much Justin.
The best lecturer evveerr💯
From Sudan 🇸🇩✌️
Was looking for it for a while.
An actual master piece, thank you so much!!
Thank you, very much sure this is a really good lecture series!
First of all, thank you so much for this amazing course. I have learned a lot from your lectures. Can I ask when this course will be updated?
Amazing lecture!
Good lecture, thanks Justin.
This is a really good lecture series!
I remember Justin from Andrej Karpathy cs231 course. It was a terrific course. Would love to check this courses again as I can see Transformer and other stuffs added here in his playlist.
hello should I see this playlist instead of cs231n? can you give your opinion please?
@@dulatormanov4809 I haven't checked the cs231n course (2017), but I only checked the one from Andrej Karpathy. Andrej is the best one, personally. I left this course halfway because I have to change my domain from ML to Full stack.
All the best for your future endeavors and good luck on your ML journey.
@@skilllessness_official thank you!!!
Thanks, it is nice lecture from Ethiopia
Thanks from Lithuania!
Great!
awesome
amazing
Great lecture, thanks for sharing. In my opinion, slide 9 should be modified: computer vision is part of machine learning. Deep learning also should be inside artificial neural network which is part of machine learning.
In my opinion, I disagree with your opinion. Let me explain why:
Computer vision tasks were performed before we start to do the same task from data (machine learning). Typical algorithms like SIFT, SURF and ORB were hand-engineered features and descriptors that were used for object recognition and matching. These didn't learn from data but were rather manually crafted. But now, we can use machine learning and mostly deep learning for computer vision since this takes away the manual form of generating features and the like. Thus, I strongly agree that CV overlaps with machine learning as depicted in the lecture slides rather than being completely part of machine learning unless your definition of being part means overlap.
For the second part of your comment, I think it will be redundant to add Deep Learning to AI if you agree that it is within ML which is already part of AI. Everything in the ML domain are automatically and completely part of the AI domain as well.
Best regards, bro.
Is there a special reason why we use pytorch instead of tensorflow(2.0) for education?? I mean for example the assignments. Is it Because its more intuitive and user friendly for beginners?
I'm a PHP web developer so is there any prerequisite knowledge that is necessary to get started with this course ? I just started watching the video course on random basis as I find it quite fascinating and interesting ...Plz let me know...
In which lecture does Prof. Justin Johnson cover autoencoders?
I'm trying to take the course, but the course homepage seems to be down. Is there any way I can get these slides?
Does this lectures has code and code exercises
Ahhh interesting
Bro,i am confused in deep learning and computer vision bcz i am not understanding when to use deep learning and when to use computer vision bcz sometimes deep learning is being used and sometimes computer vision so i am confused 😕 for what problems deep learning is used and for what problems computer vision is used bcz they are similar,i also have searched from google but not clear confusion
first go with pre requistics of computer vision and deep learning.
Ngl the history of computer vision is pretty convoluted
what is the difference between Deep Learning and Computer Vision. Are these two terms are same , technically ??
Computer vision is that you want to make the computer see
Like finger print, face recognition, classifying different images of different categories etc
Deep Learning is one of the tools to let you achieve computer vision
It’s more like looking and seeing, you look at something we involve vision , we see something we involve recognition of the vision or what we are looking at . A rope in the dark could be a snake or a rope depending on recognition, vision Is acknowledgement of an object which may be a rope or a snake . The deep learning is the intelligent differentiation between a rope or snake
😍
I strive for a future in which a ConvNet can be cognizant of why Obama stepped on that scale.
its I think better modified version of standford CS 231n ..... well done justin... but if the lecture materials and code implementations were public ,it would've been better.....
hello should I see this playlist instead of cs231n? can you give your opinion please?
@@dulatormanov4809 I think you can, he went into much more topics and explanations are quite intuitive..
@@AbidAhsan-yp4dc thank for reply! if I see this playlist it is counted that I do not skip anything from cs231n,am I right?
Amazing as they recede into history Minsky and Parpert wont be remembered for their positive contributions but more for their unintended negative impact on the field of ANN.
Can we get 1080p? This is 2020
720p is very clear
come from cs231n
I opened this video just for the vendiagram
is this still relevant in 2023?
did you find out?