Detector training

Loading...

來自 National Research University Higher School of Economics 的課程

Deep Learning in Computer Vision

90 個評分

National Research University Higher School of Economics

Deep Learning in Computer Vision

90 個評分

課程 5（共 7 門，Specialization Advanced Machine Learning）

Deep learning added a huge boost to the already rapidly developing field of computer vision. With deep learning, a lot of new applications of computer vision techniques have been introduced and are now becoming parts of our everyday lives. These include face recognition and indexing, photo stylization or machine vision in self-driving cars. The goal of this course is to introduce students to computer vision, starting from basics and then turning to more modern deep learning models. We will cover both image and video recognition, including image classification and annotation, object recognition and image search, various object detection techniques, motion estimation, object tracking in video, human action recognition, and finally image stylization, editing and new image generation. In course project, students will learn how to build face recognition and manipulation system to understand the internal mechanics of this technology, probably the most renown and oftenly demonstrated in movies and TV-shows example of computer vision and AI.

從本節課中

Object detection

In this week, we focus on the object detection task — one of the central problems in vision. We start with recalling the conventional sliding window + classifier approach culminating in Viola-Jones detector. Tracing the development of deep convolutional detectors up until recent days, we consider R-CNN and single shot detector models. Practice includes training a face detection model using a deep convolutional neural network.

Object detection problem6:15

Sliding windows3:14

HOG-based detector2:27

Detector training3:08

Viola-Jones face detector5:41

Attentional cascades and neural networks3:21

與講師見面

Anton Konushin
Senior Lecturer
HSE Faculty of Computer Science
Alexey Artemov
Senior Lecturer
HSE Faculty of Computer Science

In this video, I will talk a little bit about detector training.

How many objects and background regions are in the image?

The number of object is several magnitudes times

smaller than the number of background regions.

So, object detection is an essentially asymmetrical problem with very imbalanced classes.

The class background is very difficult,

so we need a lot of good data for training.

And good training requires similar number of objects and backgrounds examples.

I will demonstrate the training procedure using

the examples from Andrew Zisserman VGG group.

That's the training of histogram of orientation detector plus SVM for person upper torso.

The data set is created from 33 videos from Hollywood films.

Totally, 1122 frames were annotated,

with more than 1600 persons in total.

First, positive data is augmented.

Next after data augmentation for

object detector training are the same as for image classification training.

We apply random augmentations like scaling,

shifts, rotation, mirroring to positive examples.

In this particular example,

from 1600 ground truth examples,

we get 32,000 of jittered examples.

The only way to populate the negative data set,

is to use random negative examples.

As seen in this slide,

most of such examples should be very easy for the detector.

Then, we train first detector using augmented positive data and random negative data.

The first detector is applied to train images.

Our goal is to search for high-confidence false detections.

Such detections are hard examples for the detector because it fails on them.

We then add this hard-negative examples to the data set.

The classifier is then re-trained in the procedure of hard-negative mining is repeated.

Here, you can see examples for false detections.

Most of them are close to true detections but with very bad localization.

After several alteration, we plugged

the precision recall curve of the final detector against the first detector.

You can clearly see that the detector performance has improved dramatically.

So, hard examples are very important for good detector training.

What if negative data set is constructed only from hard negative examples?

It can degrade the performance.

So we usually need both easy and hard examples for training of the detector.

On this slide, the detection from

first and final detectors are demonstrated for two images.

You can see that the number of false detection has lowered.

But on the second image,

one of the true detection is missing.

Overall, hard-negative mining procedure

can greatly improve the performance of the detector.

But we should be sure that all objects has been annotated.

Otherwise, missed objects can be also detected and added to the negative examples,

and this will greatly degrade the performance of the detector.

探索我們的目錄

免費加入並獲得個性化推薦、更新和優惠。

Coursera 致力於普及全世界最好的教育，它與全球一流大學和機構合作提供在線課程。

© 2019 Coursera Inc. 保留所有權利。

通過 App Store 下載

通過 Google Play 獲取