What Is Tesla Vision? How Tesla Trains Deep Neural Nets And How You Can Learn That! (Top Courses)

Tesla has been working on something called “Tesla Vision” for quite some time now! Tesla Vision is a type of Computer Vision developed by Tesla for it’s self-driving cars. Tesla Vision is based on deep neural networks and Tesla is able to train it with the help of it’s in-house supercomputer!
There is a very important reason why Tesla is focusing on computer vision, it wants to stop relying on radar and lidar altogether! Reason being is that the biggest downside to making use of radar and lidar sensors in their cars, is the need to have extremely detailed maps of places where the cars will be operating, such as roads, lanes, alleyways, traffic stops etc. This is unfortunately not feasible in most parts of the world where Tesla would ideally want to deploy and sell its cars in. Hence it’s very important for Tesla to get rid of this market bottleneck and adopt computer vision technology fully!
Computer vision is going to play a huge role in our future and there are already so many aspects which use it, so it’s extremely important for machine learning engineers to actually specialize in this! The best way to learn is by checking out some of the specific computer vision courses that I mention in the video!
-------------------------------------------------------------------------
LINKS:
--------------------------------------------------------------------------
🖊 DOWNLOAD Machine Learning Roadmap 2021: learnml.substack.com
--------------------------------------------------------------------------
MORE VIDEOS:
--------------------------------------------------------------------------
📌I'm Starting My Machine Learning Company (Day 1)
• I'm Starting My Machin...
📌Top Machine Learning Certifications For 2021
• Top Machine Learning C...
📌Why You Should NOT Learn Machine Learning!
• Why You Should NOT Lea...
📌How I Learnt Machine Learning In 6 Steps (3 months)
• How I Learnt Machine L...
📌How To Learn Machine Learning For Free
• How To Learn Machine L...
--------------------------------------------------------------------------
Follow me:
--------------------------------------------------------------------------
Subscribe: bit.ly/subscribeToSmitha​​
LinkedIn: bit.ly/SmithaKolan​​
Instagram: bit.ly/smithacodes​​
background music: bensound.com

[@shamarbauyrzhan7997]

For the algorithm. People should know about this channel.

[@dhirajmohite5693]

Never came across this video.. Wonderful and makes me more excited towards my ML, DL, and AI journey.

[@dhirajmohite5693]

I wish you to make more videos like thos where you explain how google, Facebook, tesla are going on with their development and adoption of AI. This makes it more exciting to be committed for learning this stuff..even though if you post those videos as shorts or maybe 5 minute videos would do a great job and I feel the audience is gonna definitely like it 😀

[@hwillia204]

Excellent

[@yemyathein1443]

Awasome😍

[@FarhanHussain]

Such a great video and amazing explanation! Thank you and please keep posting more.

[@kumawatrohan]

YT algorithm recommending me good videos and channels now❤️

[@Ayuushhassan]

u r the best honestly...i hope u keep doing this and never stop..all love and respect

[@lopeznets2442]

I like your videos and they encourage me to improve my knowledge of machine learning, you have a new subscriber

[@denismatuska1981]

Great channel! Cheers from Australia :)

[@abhinav3325]

Thank you so much Mam🤗 You're videos are helping us too much.

[@TheuofGames]

thanks for the video

[@ambujsingh3844]

Smitha's video always the best... 🙌

[@TheRemiRODRIGUES]

Thank you for the video ! It's very instructive. Is self-driving car the only main motivation to learn computer vision ? As we can see in the video, computer vision is used in cinema, hospital, space, ... for image restoration, image processing, ... But many algorithms are available as easy to use features in public software as Photoshop, DaVinci, ... So it's interesting to learn how those algorithms works, but is it relevant to find professional opportunity ?

[@NimishDavidMathew]

If I do an MS in Artificial Intelligence, will I be able to learn these as part of the MS program? Or do I need to take external courses?

[@starlite5097]

Hey Smitha, thanks a lot for the video. I have to work with SLAM, drones and DL for my master's thesis and I'm curious if those courses about self driving cars would be of help to me. Keep up the good work!

[@isingcauseimliving]

One thing that differentiates cars with drones are they use the Z-axis for movement and for understanding the topography, as well. Z-axis, when the drone is placed in front of you in 2D, thus Z-axis being the direction you are viewing it from. I know I am not helping you but maybe learning about how cars move will only help you in 2 directions :) My 2 cents.

[@kewalshrestha9415]

thanks a lot 🧡.. can you answer what country are you from?

[@antoc779]

I am really starting to like this channel

[@alexeypankov8180]

can I study for a master's degree in computer vision after a bachelor's degree in mechatronics, and then work with neural networks?

[@urjasvi]

hey Smitha, thank you for the amazing explanation. Really enjoyed the video. however, If you pardon me please, I would humbly like to point out a mistake at 1:55 that map unfortunately is wrong. If you google the search string below political map of India survey of India the very first result from survey of India, with dot gov dot in domain contains the right official map of India I tried providing the link but yt deleted my comment for some reason 🙏

[@hussainshaik4390]

i like you so much :)

[@olaoluwabalogun8741]

❤❤❤

[@surflaweb]

How many GB is GPU of a tesla car?

[@krranaware8396]

Gpu is used for training model

[@surflaweb]

@@krranaware8396 And for recognition and segmentation?

[@dwivedi6037]

Ma'am are you India

[@kbmblizz1940]

Ppl who can't see the big picture are still saying, self driving will never be solved. Every day I see terrible, faulty humans 🚗, but AI must be purrfect 🐱 ?! Edit: Good point wrt developering nations, vision based is better.

[@krranaware8396]

Being in developing country India , it is not possible even with computer vision but better than radar, lidar technology. Indian roads are very very densely populated, car mirror's are almost touching bike or even people.

[@ctankep]

Automotive lidar in a manufacturing form factor : That fits into a matchbox now exceeds 20 m points / sec at 30 Hz , and that was two years ago. Absolute position beats depth estimation from 2d images and you need to then run multiple cameras. If one of those cameras gets out of alignment then its not going to estimate depth correctly. Dependence on light , bad visibility also affects computer vision. Elon just doesn't want to admit there is better , more appropriate tech than the decision " tesla " has made for better or worse. Its reflected in their inability to get to L5 self - driving this year. Even though thats what he told investors and it was a bald faced lie. The " robot " is also another classic Musk distraction to pump his stocks. There is nothing developed , and they had a person pretend to be a robot and dance around on stage , nothing to demo. Even their " self - driving " beta doesn't actually do self - driving ! Eberhard and Straubel invented and founded " tesla " then left after Musk bought in and pushed them out. The only thing he has had a hand in was the hubcap design.