brother can you make a roadmap ( plz consider this )..that 1. whay we need to study ? 2. prerequisite to learn this ? 3. resources or else what you planned for next upcoming videos ? 4. Are you going to make a proper end to end playlists for robotics ? ( plz consider this question) ROADMAP IS IMP...( if you have plz upload or else paste link if you have made already)
Thank you Raj! Good Roadmap answered on Quora: qr.ae/pGBz3y Answer to your 1st and 2nd question is in the roadmap. If you still need more details, you can email me. My email is in the About section of my TH-cam Channel. For 3rd and 4th questions, yes, I am working on building a complete end to end playlists for Robotics. As Robotics is a huge field it's going to take some time. You can subscribe to my channel as we may together explore more!
Hi, In SFA3D, is it possible to predict 3D bounding boxes on LIDAR point cloud data in place of BEV bboxes? If yes, how? If not, why? Please explain me. I am a newbie... Learning basics on 3D object detection.
Let an Indian man tell you how to solve this problem Since you can read and write English and find it hard to listen to the tutor's accent The solution is very simple: *mute the video, and turn on subtitles* You're welcome bro 😎
00:00:00 This 1-hour video tutorial provides an overview of computer vision and perception for self-driving cars, with a focus on deep learning techniques. The first project covers road segmentation using a deep learning technique called fully convolutional networks. The next project tackles 2d object detection on various traffic scenarios using a deep learning technique called the Yolo algorithm. The third project discusses multi task learning based on two tasks depth estimation and semantic segmentation. The fourth project covers 3d object detection, followed by a final project on bird's eye view visualization using an advanced computer vision technique called transformers. If you're interested in learning more about self-driving cars, this video tutorial is a good place to start. 00:05:00 Transposed convolutions are a more accurate method of up sampling than nearest neighbor methods, and they allow for more flexible weighting schemes. 00:10:00 In this video, the author discusses the architecture of a deep learning model for self-driving cars, which uses a fully convolutional network with a backbone network (VGG net). The author also discusses how to adapt the VGG net for use in this task, and how the model achieved good results on a baseline dataset. 00:15:00 The video demonstrates how to implement a deep learning model for computer vision and perception of self-driving cars. The model is trained on the image net data set, and then three blocks are extracted: pool three, pool four, and pull five layers. The first step is to apply a sample two times that is seen over here. The second step is to upsample the second two acts of sampling and add that upsampled layer with the pool three layer. The third step is to apply a convolution to D which is again a placeholder overhead. The final step is to calculate the upsampling eight times and return the model with inputs and outputs set. The model is compared to a baseline model that does not use upsampling. The results show that concatenate is also performing well, but add is a little better than this concatenate function. 00:20:00 In this video, the deep learning network trained on the Lyft data set from the 3d object detection challenge is compared with a 2d network. It is shown that the deep learning network is not getting good results, transpose convolution is not doing well, although it has learned the task. The output is pixelated and does not look good on the segmentation. 00:25:00 This video discusses the Yolo algorithm, a deep learning technique used for 2D object detection. The algorithm consists of three stages: anchor boxes, intersection over union, and bounding box predictions. In the first stage, the input image is passed through a convolutional neural network and the output is a 19 cross 19 matrix. The 19 cross 19 boxes represent what is present in the cell and are used to determine the object(s) present. The second stage calculates the position of the center of the object in relation to the cell and assigns a vector to each cell. The third stage uses the vectors to describe the object and produces a probability for each bounding box. 00:30:00 In this video, the author discusses computer vision and how it is used in self-driving cars. The author shows how intersection over union (IOU) is used to separate different bounding boxes for objects, and how the Charis YOLO v3 implementation is used to achieve better accuracy and speed. 00:35:00 In this video, the Yolo v3 object detection algorithm is used to track two cars moving in front of an autonomous car. The algorithm is explained in detail, including the Kalman filter. 00:40:00 The Kalman filter is a mathematical model used to predict the movement of objects in a video. The filter assumes a linear velocity model and generates predictions according to it. Once we get the real data where the object is, we input that again to the Kalman filter. The Kalman filter improves its predictions based on the real data that it got, and again generates a new set of predictions. The sort algorithm uses the Hungarian algorithm to solve the linear assignment problem in complexity n cubed. 00:45:00 The deep sort algorithm is used to assign detections to videos. The Yolo v3 network is used to generate bounding boxes, the detections are encoded, the Kalman filter is used to generate predictions, IOU matching is used to solve the linear assignment problem, and tracking is used to combine the different modes. 00:50:00 This video explains how computer vision and perception works for self-driving cars. The deep learning course used a pre-trained network to track objects in videos. The network was not able to correctly identify a car in a video where the car was partially obscured by another car. Increasing the weight for the Kalman filter predictions may solve the problem. 00:55:00 This video explains how computer vision and perception works in self-driving cars. The camera is on the front of the car, and the lidar is on top. The lidar generates point clouds, which are a good representation of a 3D environment. To convert between these given reference points, we use homogeneous transformations. These transformations reduce translation and rotation into a single matrix multiplication. 01:00:00 - 01:55:00 This video discusses how computer vision and perception works for self-driving cars. It explains how to convert 3D data into a 2D image using homogeneous transformations, and how to generate the LIDAR data used in self-driving cars. The video also covers the basics of the cityscapes dataset, and how to use a feature pyramid network to detect objects in an image. 01:00:00 This video explains how computer vision and perception works for self-driving cars. The video describes how to convert 3D data into a 2D image using homogeneous transformations, and how to generate the LIDAR data used in self-driving cars using this technique. 01:05:00 The cityscapes dataset consists of Street View images of cities, and the depth estimation and semantic segmentation tasks involve calculating the depth of objects in the images. The depth estimation task looks for features in the images that indicate constant depth, and the semantic segmentation task segments different objects in the images. 01:10:00 This 1-hour deep learning course discusses how to solve the multitask learning problem by using a m 10 attention network. The network is composed of a number of convolutional layers followed by a number of pooling layers. 01:15:00 The M 10 network is a deep learning technique that was designed specifically for self-driving cars. This network is composed of several layers of convolution, a merge filter, and an attention module. The network is trained to recognize different objects in a scene and predicts their locations. 01:20:00 This 1-minute video explains the use of deep learning for self-driving cars. The video shows how a computer vision and perception network is used to detect and classify objects in a live camera feed. 01:25:00 Feature extraction is the process of extracting features from an image. Feature pyramid networks are a general technique that can be employed on a number of different neural networks, including a ResNet network, to generate feature maps. These feature maps can be used to perform tasks like object detection. 01:30:00 This 1-minute video explains how computer vision and perception works for self-driving cars. The main components of the system are a deep learning network, a focal loss, and distance and direction measurements. The heat map and confidence matrix show the network's classification accuracy. The L1 and Elven losses penalize the network for making mistakes in its predictions, while the balanced Elvan loss encourages the network to learn accurate predictions. 01:35:00 The SFA 3d technique uses three different loss functions to train a neural network to detect 3d objects in a scene. The network is able to detect cars, cyclists, and pedestrians in the scene, and the final output is displayed in a video that is slowed down to show key observations. 01:40:00 The video discusses computer vision and perception for self-driving cars, and shows how to calculate a bird's eye view of the environment. The data set used in the video is derived from a simulation of a self-driving car driving around a simulated environment. 01:45:00 In this video, the unit access T technique for self-driving cars will be discussed. The unit network unit architecture and its extension, the Xs D, are used to derive a bird's eye view image from an image. The transformation between two images is called homography and it is a simple matrix multiplication. This is why the output of the unit access T is not integer values, and a lot of averaging is needed to make the final image look good. 01:50:00 In this video, the author explains the process of inverse homography, which is used to map an output pixel to an input pixel in a rotated fashion. The inverse homography calculation is done using a technique called linear interpolation. This process is then applied to a unit xs D technique for self-driving cars, where the spatial transformer learns the homography matrix parameters in order to aid the convolution network. 01:55:00 This video discusses the basics of computer vision and perception for self-driving cars. It covers the theory behind the process, including how feature maps are generated and combined using spatial transformers. The video then goes on to show how this process is implemented in code, and how results can be seen in practice by training a network on a dataset of images from a self-driving car.
My crypto mentor Mr Edward Jones, you may have come across him on a few interviews I invested $5000 last three weeks and it profited me $18,620 a higher success
Wow I know Mr Edward Jones. I met him at a conference in The Dome at America’scenter Life 2016 where he introduced us his business strategy, he helped me cover my Loans
I also trade with Mr Edward , and I must say he makes money making seem a whole lot easier right now I’m a single parent and I pay the bills comfortably since I met Edward Jones he’s absolutely amazing and I recommend him the best
I was thinking of doing a project on this 2 days back scouring all over the place for resources. You guys resolved 2 weeks worth of research in a 2 hr video. Amazing.
Okay I'm completely here by accident and have absolutely no interest in learning this (I don't have a reason to) so can I just ask why the hell the rest of you are here? What "project" are you doing that requires understanding something like this? I mean seriously coming to these comment sections is depressing, i feel like someone who has accomplished absolutely nothing, but I want to do something. Any advice or insight you have would be great
@@Apathy474 when this came out I was an undergraduate student in India studying at an university which was not among the best in the country. That meant the best opportunities in my career would be harder to achieve as compared to someone in the top rung unis. (For example, a position at Google or admission into prestigious graduate programs). I did a project for an ug course after understanding how monocular to stereo worked from this video and further research and kept on working towards building a profile as an enthusiast of deep learning. Fast forward 2 years or so, I'm a student at KTH in Sweden, a top 100 University in the world, studying what I love. I guess things worked out for me in the long run. And as long as you consistently work towards something, be it whatever, you'll surely see some favorable results as well! More power to you ✨
KITTI 3D Data Visualization | Homogenous Transformations notebook is noteworking TypeError: SetColor argument 1: expected a sequence of 3 values, got 18347 values 2nd cell in the visualization block index_list = random.sample(range(len(dataset)), 5) images = [] for index in index_list: visualize_points(index) images.append(mpimg.imread(f"point_image_{index}.png")) #can anyone help
I have downloaded the datasets and created the environment, however there is not a single indication that where are the images coming from! where to upload the image. I have a basic understanding but this tutorials has to be more clear! I keep running my code and it keeps throwing errors. Thank you
Hi, thanks for this helpful video. I would like to ask you if I want be in this level in deep learning and computer vision, machine learning expert from where should I start learning? Specially learning coding or I must start from full-stack development then learning python?
Hello Yousef! I think you should start with learning Python, and do some simple projects in Python. Once you have a decent knowledge of Python, then start learning Machine Learning basics. After Machine Learning you can jump to Computer Vision and Deep Learning.
To all the Indians who are watching this video trust me as a person studying in abroad at a prestigious institute I felt like we were taught almost the same content. Instant of searching for jobs there is a huge requirement to begin some startups in Indian. Now you have knowledge about basic level self driving! please start puting the ideas into building small product based companies. I am telling the western people are intellectual but not as innovative as Indians and all they want is to mock Indians and work for them. Please please please start making a difference.
Thank you. I was not a great scholar but I have an idea I am trying to create. I am not indian but I will take this critic as hope this small idea can change things.
I really learn a lot by this course. It was a bunch of knowledge delivered in just 2 hours of this course. It took me 1 complete day to learn with managing notes for my self. I had no knowledge about computer vision with this self driving cars now I have much to continue this field further. Thank You very much for this course. 😍
It was really hard to follow for me at least due to the strong accent but hopefully others could follow along. He seems very knowledgeable and thanks for the vid!
Another is the sensor of "auto brake" on normal vehicles to prevent accidents, to interrupt the brake point to just brake safely if have vehicles, even avoiding on left and right or speeding if from behind and safe in front.
Well at least writing back-end code for billing will not be sued, but perhaps a driving program goes bad and causes someone harm and they file lawsuit.
Hahahaha, will not be what? Make a mistake on double vs float, calculate wrong taxes and you are going out of business faster than you can say "backend". :D
@@justwanderin847 In billing not, but if you extend your example to any kind of standard software - f***ing up a simple table view within pharma environment, you are equally effed. But then again thats why you need to get your Autopilot certified and your instabookfacegram clone not ^^
@@unknownguywholovespizza i know ds and algos little bit, but i am very good at C, and also am trying to good at cpp and dp, but fun fact, i am student of electronic engineer, so i need to learn assembly, also my goal is to get embedded system design course at 3th year
Thanks a lot for featuring the course ☺.
Hope you all learn something new from this.
All the feedback is greatly appreciated!
Sakshay brother..quality content....i love white board detailed videos.........
brother can you make a roadmap ( plz consider this )..that
1. whay we need to study ?
2. prerequisite to learn this ?
3. resources or else what you planned for next upcoming videos ?
4. Are you going to make a proper end to end playlists for robotics ? ( plz consider this question)
ROADMAP IS IMP...( if you have plz upload or else paste link if you have made already)
Thank you Raj!
Good Roadmap answered on Quora: qr.ae/pGBz3y
Answer to your 1st and 2nd question is in the roadmap. If you still need more details, you can email me. My email is in the About section of my TH-cam Channel.
For 3rd and 4th questions, yes, I am working on building a complete end to end playlists for Robotics. As Robotics is a huge field it's going to take some time. You can subscribe to my channel as we may together explore more!
@@RoboticswithSakshay ❤
How about a similar thing for aircraft like drones? Thanks for the great work!
It is just unbelievable the amount of knowledge one can certainly acquire through your amazing courses.
Thats the big fallacy, you get demonstrated a few core concepts, but you do not at all acquire knowledge. You are barely scratching the surface.
@@StarryNightSky587 but if it's working its good
- Some guy
@@StarryNightSky587 And that sir, is also completely accurate.
@@StarryNightSky587 That's good though, that's how you know if you want to continue with it, and if you do then you need to do a deepdive anyway.
Exactly what I needed !
Oh, my SFA3D project is introduced in the video😊
well done my friend !
Great course
Una joyita!!!
Hi, In SFA3D, is it possible to predict 3D bounding boxes on LIDAR point cloud data in place of BEV bboxes? If yes, how? If not, why? Please explain me. I am a newbie... Learning basics on 3D object detection.
I am doing similar thing, but with VELODYNE VLP-16 dataset. Can you please point of any references/codes regarding that ?
thank god this video has subtitle
thank you for saving my a**
Indian
Hate to say it but the Indian accent sounds so annoying.
Which country are you from?
Hate to say it but your comment is annoying
May I tell you an annoying thing, he is my classmate and is a currently a Btech Student 😅
Let an Indian man tell you how to solve this problem
Since you can read and write English and find it hard to listen to the tutor's accent
The solution is very simple:
*mute the video, and turn on subtitles*
You're welcome bro 😎
00:00:00 This 1-hour video tutorial provides an overview of computer vision and perception for self-driving cars, with a focus on deep learning techniques. The first project covers road segmentation using a deep learning technique called fully convolutional networks. The next project tackles 2d object detection on various traffic scenarios using a deep learning technique called the Yolo algorithm. The third project discusses multi task learning based on two tasks depth estimation and semantic segmentation. The fourth project covers 3d object detection, followed by a final project on bird's eye view visualization using an advanced computer vision technique called transformers. If you're interested in learning more about self-driving cars, this video tutorial is a good place to start.
00:05:00 Transposed convolutions are a more accurate method of up sampling than nearest neighbor methods, and they allow for more flexible weighting schemes.
00:10:00 In this video, the author discusses the architecture of a deep learning model for self-driving cars, which uses a fully convolutional network with a backbone network (VGG net). The author also discusses how to adapt the VGG net for use in this task, and how the model achieved good results on a baseline dataset.
00:15:00 The video demonstrates how to implement a deep learning model for computer vision and perception of self-driving cars. The model is trained on the image net data set, and then three blocks are extracted: pool three, pool four, and pull five layers. The first step is to apply a sample two times that is seen over here. The second step is to upsample the second two acts of sampling and add that upsampled layer with the pool three layer. The third step is to apply a convolution to D which is again a placeholder overhead. The final step is to calculate the upsampling eight times and return the model with inputs and outputs set. The model is compared to a baseline model that does not use upsampling. The results show that concatenate is also performing well, but add is a little better than this concatenate function.
00:20:00 In this video, the deep learning network trained on the Lyft data set from the 3d object detection challenge is compared with a 2d network. It is shown that the deep learning network is not getting good results, transpose convolution is not doing well, although it has learned the task. The output is pixelated and does not look good on the segmentation.
00:25:00 This video discusses the Yolo algorithm, a deep learning technique used for 2D object detection. The algorithm consists of three stages: anchor boxes, intersection over union, and bounding box predictions. In the first stage, the input image is passed through a convolutional neural network and the output is a 19 cross 19 matrix. The 19 cross 19 boxes represent what is present in the cell and are used to determine the object(s) present. The second stage calculates the position of the center of the object in relation to the cell and assigns a vector to each cell. The third stage uses the vectors to describe the object and produces a probability for each bounding box.
00:30:00 In this video, the author discusses computer vision and how it is used in self-driving cars. The author shows how intersection over union (IOU) is used to separate different bounding boxes for objects, and how the Charis YOLO v3 implementation is used to achieve better accuracy and speed.
00:35:00 In this video, the Yolo v3 object detection algorithm is used to track two cars moving in front of an autonomous car. The algorithm is explained in detail, including the Kalman filter.
00:40:00 The Kalman filter is a mathematical model used to predict the movement of objects in a video. The filter assumes a linear velocity model and generates predictions according to it. Once we get the real data where the object is, we input that again to the Kalman filter. The Kalman filter improves its predictions based on the real data that it got, and again generates a new set of predictions. The sort algorithm uses the Hungarian algorithm to solve the linear assignment problem in complexity n cubed.
00:45:00 The deep sort algorithm is used to assign detections to videos. The Yolo v3 network is used to generate bounding boxes, the detections are encoded, the Kalman filter is used to generate predictions, IOU matching is used to solve the linear assignment problem, and tracking is used to combine the different modes.
00:50:00 This video explains how computer vision and perception works for self-driving cars. The deep learning course used a pre-trained network to track objects in videos. The network was not able to correctly identify a car in a video where the car was partially obscured by another car. Increasing the weight for the Kalman filter predictions may solve the problem.
00:55:00 This video explains how computer vision and perception works in self-driving cars. The camera is on the front of the car, and the lidar is on top. The lidar generates point clouds, which are a good representation of a 3D environment. To convert between these given reference points, we use homogeneous transformations. These transformations reduce translation and rotation into a single matrix multiplication.
01:00:00 - 01:55:00
This video discusses how computer vision and perception works for self-driving cars. It explains how to convert 3D data into a 2D image using homogeneous transformations, and how to generate the LIDAR data used in self-driving cars. The video also covers the basics of the cityscapes dataset, and how to use a feature pyramid network to detect objects in an image.
01:00:00 This video explains how computer vision and perception works for self-driving cars. The video describes how to convert 3D data into a 2D image using homogeneous transformations, and how to generate the LIDAR data used in self-driving cars using this technique.
01:05:00 The cityscapes dataset consists of Street View images of cities, and the depth estimation and semantic segmentation tasks involve calculating the depth of objects in the images. The depth estimation task looks for features in the images that indicate constant depth, and the semantic segmentation task segments different objects in the images.
01:10:00 This 1-hour deep learning course discusses how to solve the multitask learning problem by using a m 10 attention network. The network is composed of a number of convolutional layers followed by a number of pooling layers.
01:15:00 The M 10 network is a deep learning technique that was designed specifically for self-driving cars. This network is composed of several layers of convolution, a merge filter, and an attention module. The network is trained to recognize different objects in a scene and predicts their locations.
01:20:00 This 1-minute video explains the use of deep learning for self-driving cars. The video shows how a computer vision and perception network is used to detect and classify objects in a live camera feed.
01:25:00 Feature extraction is the process of extracting features from an image. Feature pyramid networks are a general technique that can be employed on a number of different neural networks, including a ResNet network, to generate feature maps. These feature maps can be used to perform tasks like object detection.
01:30:00 This 1-minute video explains how computer vision and perception works for self-driving cars. The main components of the system are a deep learning network, a focal loss, and distance and direction measurements. The heat map and confidence matrix show the network's classification accuracy. The L1 and Elven losses penalize the network for making mistakes in its predictions, while the balanced Elvan loss encourages the network to learn accurate predictions.
01:35:00 The SFA 3d technique uses three different loss functions to train a neural network to detect 3d objects in a scene. The network is able to detect cars, cyclists, and pedestrians in the scene, and the final output is displayed in a video that is slowed down to show key observations.
01:40:00 The video discusses computer vision and perception for self-driving cars, and shows how to calculate a bird's eye view of the environment. The data set used in the video is derived from a simulation of a self-driving car driving around a simulated environment.
01:45:00 In this video, the unit access T technique for self-driving cars will be discussed. The unit network unit architecture and its extension, the Xs D, are used to derive a bird's eye view image from an image. The transformation between two images is called homography and it is a simple matrix multiplication. This is why the output of the unit access T is not integer values, and a lot of averaging is needed to make the final image look good.
01:50:00 In this video, the author explains the process of inverse homography, which is used to map an output pixel to an input pixel in a rotated fashion. The inverse homography calculation is done using a technique called linear interpolation. This process is then applied to a unit xs D technique for self-driving cars, where the spatial transformer learns the homography matrix parameters in order to aid the convolution network.
01:55:00 This video discusses the basics of computer vision and perception for self-driving cars. It covers the theory behind the process, including how feature maps are generated and combined using spatial transformers. The video then goes on to show how this process is implemented in code, and how results can be seen in practice by training a network on a dataset of images from a self-driving car.
Thank you man
Bless you man. This was the comment I was looking for!!
Investing in different streams of income in other not to depend on government for funds and avoid all the chitchat about the inflation bla bla bla
Interesting. I have a lump sum doing absolutely nothing at all in my bank account, I wanna get something started with it, any reasonable ideal
There are platform where you can invest and they trade your money. Then pay you profit either weekly or monthly. That's investing.
My crypto mentor Mr Edward Jones, you may have come across him on a few interviews I invested $5000 last three weeks and it profited me $18,620 a higher success
Wow I know Mr Edward Jones. I met him at a conference in The Dome at America’scenter Life 2016 where he introduced us his business strategy, he helped me cover my Loans
I also trade with Mr Edward , and I must say he makes money making seem a whole lot easier right now I’m a single parent and I pay the bills comfortably since I met Edward Jones he’s absolutely amazing and I recommend him the best
What a timing. Ouf. Needed this for my Drone project seriously
This channel is such a treasure for engineers... The title of video surprises me a lot, just wow.
you guys have been killing it, thank you
I was thinking of doing a project on this 2 days back scouring all over the place for resources. You guys resolved 2 weeks worth of research in a 2 hr video. Amazing.
Okay I'm completely here by accident and have absolutely no interest in learning this (I don't have a reason to) so can I just ask why the hell the rest of you are here? What "project" are you doing that requires understanding something like this? I mean seriously coming to these comment sections is depressing, i feel like someone who has accomplished absolutely nothing, but I want to do something. Any advice or insight you have would be great
@@Apathy474 when this came out I was an undergraduate student in India studying at an university which was not among the best in the country. That meant the best opportunities in my career would be harder to achieve as compared to someone in the top rung unis. (For example, a position at Google or admission into prestigious graduate programs). I did a project for an ug course after understanding how monocular to stereo worked from this video and further research and kept on working towards building a profile as an enthusiast of deep learning. Fast forward 2 years or so, I'm a student at KTH in Sweden, a top 100 University in the world, studying what I love. I guess things worked out for me in the long run. And as long as you consistently work towards something, be it whatever, you'll surely see some favorable results as well! More power to you ✨
KITTI 3D Data Visualization | Homogenous Transformations notebook is noteworking
TypeError: SetColor argument 1: expected a sequence of 3 values, got 18347 values
2nd cell in the visualization block
index_list = random.sample(range(len(dataset)), 5)
images = []
for index in index_list:
visualize_points(index)
images.append(mpimg.imread(f"point_image_{index}.png"))
#can anyone help
How about a similar thing for aircraft like drones? Thanks for the great work!
There are already many unmanned planes.
I have downloaded the datasets and created the environment, however there is not a single indication that where are the images coming from! where to upload the image. I have a basic understanding but this tutorials has to be more clear! I keep running my code and it keeps throwing errors. Thank you
Hi, thanks for this helpful video.
I would like to ask you if I want be in this level in deep learning and computer vision, machine learning expert from where should I start learning? Specially learning coding or I must start from full-stack development then learning python?
Hello Yousef!
I think you should start with learning Python, and do some simple projects in Python. Once you have a decent knowledge of Python, then start learning Machine Learning basics. After Machine Learning you can jump to Computer Vision and Deep Learning.
@@RoboticswithSakshay Thank you so much
To all the Indians who are watching this video trust me as a person studying in abroad at a prestigious institute I felt like we were taught almost the same content. Instant of searching for jobs there is a huge requirement to begin some startups in Indian. Now you have knowledge about basic level self driving! please start puting the ideas into building small product based companies. I am telling the western people are intellectual but not as innovative as Indians and all they want is to mock Indians and work for them. Please please please start making a difference.
Thank you. I was not a great scholar but I have an idea I am trying to create. I am not indian but I will take this critic as hope this small idea can change things.
problem is converting this theoretical knowledge into practical one
Bhai ki smile ,
Savage
I love indian man
Dude this is our project of 9 months (IT project)
we need a model to fix the sound echo issue in the video :)😃
let's have a course to use the segmented image to autonomously drive the car.
can you make a video on state estimation & localization using extended kalman filter. Please
Thanks a lot...Please add more on self Driving car.
Thanks 😄😄🎊🎊🎉🎊🎉
Sakshay brother..quality content....i love white board detailed videos.........
Sir please upload SSIS complete video
Tesla Vision is amazing....
i have been waiting for this
Mann this channel is amazing!!!!
I really learn a lot by this course. It was a bunch of knowledge delivered in just 2 hours of this course. It took me 1 complete day to learn with managing notes for my self. I had no knowledge about computer vision with this self driving cars now I have much to continue this field further. Thank You very much for this course. 😍
Thanks! It's Amazing
😢o
Tttoooppp
Are you planning on making a full course of Trigonometry?
Gilfoyle
ЭТО очень отличный apk и МНЕ ЭТО НРАВИТСЯ
Thanks
Awsome.
It was really hard to follow for me at least due to the strong accent but hopefully others could follow along. He seems very knowledgeable and thanks for the vid!
Commenting for algorythm thanks.
👍🏾
I feel of having the turning identification on an Electric Skateboard(control by 1 Hand if "Manual Turning") "to follow road/path".
Electronic*
Another is the sensor of "auto brake" on normal vehicles to prevent accidents, to interrupt the brake point to just brake safely if have vehicles, even avoiding on left and right or speeding if from behind and safe in front.
awesome content! big thanks
This is what i wanted from so long
Thnx a lot. Very useful chanel.
Thank you for this course!!
Wow! I 'll try it as soon as possible.
Is codecamp ai course worth it?
Impressive details and quality 🎉
pom pom pom pom :D :D :D
Surprising tat some such is made freely available
Great
Watch the video before commenting bruh😂
ddddddd
ooh la la
ХОРОШО
Bad kick
Well at least writing back-end code for billing will not be sued, but perhaps a driving program goes bad and causes someone harm and they file lawsuit.
Hahahaha, will not be what? Make a mistake on double vs float, calculate wrong taxes and you are going out of business faster than you can say "backend". :D
@@StarryNightSky587 True, but no one will be hospitalized or worse.
@@justwanderin847 In billing not, but if you extend your example to any kind of standard software - f***ing up a simple table view within pharma environment, you are equally effed.
But then again thats why you need to get your Autopilot certified and your instabookfacegram clone not ^^
Quw chingon
ASSEMBLY FULL COURSE PLEASE
No not that one plz 😱
@@unknownguywholovespizza i know ds and algos little bit, but i am very good at C, and also am trying to good at cpp and dp, but fun fact, i am student of electronic engineer, so i need to learn assembly, also my goal is to get embedded system design course at 3th year
👍👍
Indian again...
How to create our own programming language from scratch with python
Da hell
Dayum! That's new
This, without python
Build your own compiler, start reading the Dragon's book
@@breno9138 How can I get that online for free?
I wish I can understand indian