Download "Deep Learning Anomaly Detection with MVTec MERLIC"

Please wait. We're preparing links for easy ad-free video watching and downloading.

Previous video: What's the difference between a Point and A Pip?Next video: MILF BODY TYPE WITH BASIC SHAPES (How To Draw)

0:00

Introduction

0:38

What is Anomaly Detection?

1:56

Example “detect_anomalies_of_bottles.mvapp”

2:56

Training with only “good” images

4:00

Results inside the Creator

4:34

Training with the addition of “bad” images

5:25

Visualization within the Frontend

6:11

Outro

MILF BODY TYPE WITH BASIC SHAPES (How To Draw)

MILF BODY TYPE WITH BASIC SHAPES (How To Draw)

Channel: mikeymegamega

What's the difference between a Point and A Pip?

What's the difference between a Point and A Pip?

Channel: Jimdandy1958

C4D TUTORIAL | Low Poly Apartment & Shop (Beginner)

C4D TUTORIAL | Low Poly Apartment & Shop (Beginner)

Channel: Twistereli

Tutorial: Exotic Pole Dance Intermediate Trick - Pivot Down Transition - Lara Joh

Tutorial: Exotic Pole Dance Intermediate Trick - Pivot Down Transition - Lara Joh

Channel: Lara JPole

Моделирование дивана в Blender 2.8 - Часть 6 | Создание швов | Уроки на русском

Моделирование дивана в Blender 2.8 - Часть 6 | Создание швов | Уроки на русском

Channel: Denis Kozhar

Animating Environments with Depth Maps - Lazy Tutorials

Animating Environments with Depth Maps - Lazy Tutorials

Channel: IanHubert

Элиезер Юдковский про ИИ, клубнику, банк спермы, режим бога и лучших ученых на острове, 20.02.2023

Элиезер Юдковский про ИИ, клубнику, банк спермы, режим бога и лучших ученых на острове, 20.02.2023

ICT13 AI 07 Software Agent Roshan Rolanka

ICT13 AI 07 Software Agent Roshan Rolanka

Channel: ICT rolanka

黔驴技穷？Photoshop 2023都挤了些啥？

黔驴技穷？Photoshop 2023都挤了些啥？

Channel: 巫师后期.wrcolor

Your First 2D GAME From Zero with GODOT 4! **Vampire Survivor Style**

Your First 2D GAME From Zero with GODOT 4! Vampire Survivor Style

Channel: GDQuest

MERLIC

5.0

MVTec

AI

Artificial Intelligence

KI

Künstliche Intelligenz

Deep Learning

DL

Anomaly Detection

AD

tutorial

training

introduction

beginner

screencast

machine vision

Bildverarbeitung

Maschinensehen

Maschinelles Sehen

00:00:07

hello and welcome

00:00:09

in this video i will introduce you to

00:00:11

mvtech merlick's deep learning based

00:00:13

detect anomalies tool

00:00:18

for this we are taking a closer look at

00:00:20

the myrlic example detect anomalies of

00:00:23

bottles.mv app

00:00:30

the goal of this example application is

00:00:32

to inspect glass bottles and detect

00:00:34

defects like these

00:00:38

but first

00:00:40

what is anomaly detection

00:00:43

anomaly detection at its core is a

00:00:45

binary classification which means an

00:00:48

input gets assigned to either one of two

00:00:51

classes

00:00:53

you input an image and the anomaly

00:00:55

detection will tell you if it's good or

00:00:58

bad

00:01:01

it's based on unsupervised learning only

00:01:04

one of the two classes must be trained

00:01:07

the other one is an implicit rejection

00:01:10

class therefore elaborate labeling is not

00:01:13

needed

00:01:16

the trained class is referred to as good

00:01:18

and the implicit rejection class as bad

00:01:24

the use case of anomaly detection in

00:01:26

merlick is to find local anomalies

00:01:32

it is especially useful for applications

00:01:34

where you have many good samples but the

00:01:36

defects are rare and diverse

00:01:41

anomaly detection requires only a small

00:01:44

number of images to train its model in

00:01:46

some cases as few as 20 images are

00:01:48

sufficient this makes anomaly detection

00:01:50

easy to train and quick to prototype

00:01:57

the image source tool is for the image

00:01:59

acquisition the tools determine

00:02:02

alignment with matching and align image

00:02:04

are used to pre-process the images for

00:02:06

the detect anomalies tool

00:02:09

the detect anomalies tool performs the

00:02:11

anomaly detection

00:02:13

note it's quick info where you can find

00:02:15

an overview on how to use the detect

00:02:17

anomalies tool

00:02:19

merlick supports both cpus and gpus as

00:02:22

processing devices

00:02:24

for this tutorial i changed up the

00:02:26

example in a minor way

00:02:28

these changes were made to show you how

00:02:30

to train and apply an anomaly detection

00:02:32

model within mv tech merlick i deleted

00:02:35

the already trained anomaly detection

00:02:37

model and configured the image source

00:02:39

tool for the acquisition of the training

00:02:41

data

00:02:43

additionally i split the training data

00:02:45

into good and bad images

00:02:48

bad images aren't needed for training

00:02:50

but they can significantly improve the

00:02:52

result of the anomaly detection

00:02:56

now let's start the configuration of the

00:02:58

detect anomalies tool first we must

00:03:01

train our anomaly detection model for

00:03:03

this we must select the folder

00:03:05

containing the good training data within

00:03:07

the image source tool we start our

00:03:10

training by going into the detect

00:03:11

anomalies tool and adding the data one

00:03:14

by one to our training you can either

00:03:16

click the plus button right here or

00:03:19

press the f3 key on your keyboard and

00:03:21

label it right after you go through your

00:03:23

data by clicking execute merlivision app

00:03:26

once or by pressing f6 on your keyboard

00:03:30

do this for every image until you

00:03:32

encounter images you've already inserted

00:03:33

into the training data you've already

00:03:36

added to the training is marked as

00:03:37

inserted

00:03:44

duplicated or unwanted data can be

00:03:46

deleted from the training via the x

00:03:48

button

00:03:50

after successfully adding all images to

00:03:52

the training set we can train our

00:03:54

anomaly detection model this might take

00:03:56

a while

00:04:00

for the inference we change our image

00:04:02

source to the folder containing the

00:04:04

process data

00:04:06

in the detect anomalies tool you can now

00:04:08

see the processing of the images as a

00:04:11

result you get a heat map and the

00:04:13

boolean anomaly detected in the bottom

00:04:15

left corner

00:04:19

here you can also step through the

00:04:20

inference images one by one clicking

00:04:23

execute merlivision app once or pressing

00:04:26

f6 on your keyboard

00:04:29

as you can see the application is

00:04:31

already pretty good at detecting

00:04:33

anomalies

00:04:35

but to make it even better we change our

00:04:37

image source to the folder containing

00:04:39

our bad training data

00:04:42

we go back into the anomaly detection

00:04:44

tool and start adding our bad images to

00:04:47

the training set just like we learned

00:04:49

before

00:05:00

afterwards we train our anomaly

00:05:01

detection model once more

00:05:09

these bad images are technically not

00:05:11

used for training but to evaluate the

00:05:14

anomaly model which results in better

00:05:16

adapted anomaly thresholds this means

00:05:18

that the anomaly detection is better at

00:05:20

deciding whether images contain

00:05:22

anomalies or not

00:05:25

for a final look at our application we

00:05:27

change our image source back to the

00:05:29

folder containing the processing data

00:05:35

to better visualize our program i set up

00:05:37

a small front end exactly like the one

00:05:40

inside the example detect anomalies of

00:05:42

bottles.mv app

00:05:46

as we look at our front end you can

00:05:48

clearly see which images contain

00:05:50

anomalies

00:05:51

anomalies are indicated by a red light

00:05:55

additionally the heat map shows where

00:05:57

exactly those anomalies are

00:06:01

the applications of anomaly detection

00:06:03

are countless other examples would be

00:06:05

the inspection of solder joints or

00:06:07

checking for defects in wood

00:06:12

this concludes this video you should now

00:06:15

be able to navigate through the example

00:06:17

detect anomalies of bottles.mv app and

00:06:20

adapt it according to your application

00:06:24

if you have feedback for us regarding

00:06:26

merlick feel free to send it to us via

00:06:28

the feedback button in the top right

00:06:30

corner

00:06:31

thank you for watching

Description:

In this tutorial, you will learn how to use MVTec MERLIC’s new “Detect Anomalies”-Tool. We will take a look at the example application “detect_anomalies_of_bottles.mvapp” and you will learn what anomaly detection actually is, and what kinds of applications it can be used for. Content 0:00 Introduction 0:38 What is Anomaly Detection? 1:56 Example “detect_anomalies_of_bottles.mvapp” 2:56 Training with only “good” images 4:00 Results inside the Creator 4:34 Training with the addition of “bad” images 5:25 Visualization within the Frontend 6:11 Outro In this video, MERLIC 5.0 is used. https://www.mvtec.com/ https://www.mvtec.com/products/merlic

Preparing download options

Popular

HD video

Only sound

All

* — If the video is playing in a new tab, go to it, then right-click on the video and select "Save video as..."

** — Link intended for online playback in specialized players

Questions about downloading video

How can I download "Deep Learning Anomaly Detection with MVTec MERLIC" video?

http://unidownloader.com/ website is the best way to download a video or a separate audio track if you want to do without installing programs and extensions.
The UDL Helper extension is a convenient button that is seamlessly integrated into YouTube, Instagram and OK.ru sites for fast content download.
UDL Client program (for Windows) is the most powerful solution that supports more than 900 websites, social networks and video hosting sites, as well as any video quality that is available in the source.
UDL Lite is a really convenient way to access a website from your mobile device. With its help, you can easily download videos directly to your smartphone.

Which format of "Deep Learning Anomaly Detection with MVTec MERLIC" video should I choose?

The best quality formats are FullHD (1080p), 2K (1440p), 4K (2160p) and 8K (4320p). The higher the resolution of your screen, the higher the video quality should be. However, there are other factors to consider: download speed, amount of free space, and device performance during playback.

Why does my computer freeze when loading a "Deep Learning Anomaly Detection with MVTec MERLIC" video?

The browser/computer should not freeze completely! If this happens, please report it with a link to the video. Sometimes videos cannot be downloaded directly in a suitable format, so we have added the ability to convert the file to the desired format. In some cases, this process may actively use computer resources.

How can I download "Deep Learning Anomaly Detection with MVTec MERLIC" video to my phone?

You can download a video to your smartphone using the website or the PWA application UDL Lite. It is also possible to send a download link via QR code using the UDL Helper extension.

How can I download an audio track (music) to MP3 "Deep Learning Anomaly Detection with MVTec MERLIC"?

The most convenient way is to use the UDL Client program, which supports converting video to MP3 format. In some cases, MP3 can also be downloaded through the UDL Helper extension.

How can I save a frame from a video "Deep Learning Anomaly Detection with MVTec MERLIC"?

This feature is available in the UDL Helper extension. Make sure that "Show the video snapshot button" is checked in the settings. A camera icon should appear in the lower right corner of the player to the left of the "Settings" icon. When you click on it, the current frame from the video will be saved to your computer in JPEG format.

What's the price of all this stuff?

It costs nothing. Our services are absolutely free for all users. There are no PRO subscriptions, no restrictions on the number or maximum length of downloaded videos.