How Neural Nets estimate depth from 2D images? Monocular Depth Estimation Explained!

preview_player
Показать описание
In this video, we will be discussing the MiDAS paper, Depth Anything V1, and the latest Depth Anything V2 paper! We are going to learn the basics of Monocular Depth Estimation and some of the modern tricks, datasets, networks, and loss functions used to train these models.

Members will get access the full write-up, all the animations, powerpoint slides, and annotated PDFs from this video for channel supporters and Patrons!

Visit the Patreon link to see what is available:
#science #machinelearning #deeplearning

Relevant videos:

Links:

Timestamps:
0:00 - Intro
1:20 - MiDAS
3:05 - Depth Anything V1
4:28 - Disparity Space
5:13 - Scale and Shift Invariant Loss
5:56 - Gradient Matching Loss
8:13 - Cut Mix Augmentation
8:30 - Semantic Assisted Perception
9:46 - Synthetic Datasets
11:20 - Depth Anything V2
Рекомендации по теме
Комментарии
Автор

Best explanation of monocular depth estimation i have seen so far

vishalchahar
Автор

Thanks for your great video. I also enjoyed reading your MDE article on medium!

borhanpetgar
Автор

Great piece, thanks for the effort you put into it.

manmj
Автор

Thank you for the great content! There is no training script in the paper's GitHub repository. Could you make a video or write an article on Medium providing training code for the Depth Anything v2 ? Thank you.

Eucalyptus_Study
Автор

You are a fantastic explainer. Do you teach a course online ?
Liked & Subscribed
cheers
b

behrampatel
Автор

Bro, bro, , bro, read carefully

Thanks A LOT FOR THIS FUCKING AWESOME

I enjoy all of your. videos, i combine your videos, with my university classes and some books and im learning at fucking scary pace, thanks for all of this videos, this helps me a lot for being a good engineer in this field. PD: do you have any book recommendation to read about computer vision ?

octaviusp