代做ECS797P/U Machine Learning for Visual Data Analysis Main Examination Period 2022 Semester B代写Java程

Main Examination Period 2022 Semester B

ECS797P/U Machine Learning for Visual Data Analysis

Question 1

a)  In Figure 1, we depict an image and the plot of the entropy of the intensity in a disk around a pixel A, as a function of the radius of the disk.

i)   Make similar plots of the entropy around the pixels B and C with a brief explanation about the shape of the plot.

ii)  Explain which of the points A, B, and C, will be selected as salient points, and at which scale.

 

Figure 1

[9 marks]

b)  Explain briefly the differences between Action Recognition, Temporal Action

Localisation and Spatiotemporal Action Localisation. What is the input and what is the output of an algorithm in each case?

[6 marks]

c)  A computer vision company designs an action recognition system that aims at counting how many times customers perform. the action “drinking” in a pub setting, such as in Figure 2.

i)   Explain, why a part-based localisation method is a good choice in this setting.

ii)  What kind of descriptors would you recommend to be extracted in the spatiotemporal interest point regions and why?

iii)  Suggest how the developer of the system should choose the number of clusters in their visual codebook.

 

Figure 2

[8 marks]

Question 2

a)  Describe what do we mean by the term optical flow field and whether it is the same as the true motion field. What is the input and what is the output of an optical flow algorithm?

[6 marks]

b)  Explain why the optical flow equation is not sufficient by itself to estimate an optical flow field. How can one overcome this problem?

[5 marks]

c)  Explain how the optical flow equation is used in the Lucas-Kanade method for optical flow estimation.

[4 marks]

d)  Some Neural Network (NN) based methods for Optical Flow estimation, such as

FlowNet, use a network to predict the optical flow field. Describe the first and the last layer of such networks. Describe what is the supervisory signal (or optimisation criterion), how it is obtained and state whether it is an instance of supervised or unsupervised Machine Learning.

[10 marks]


Question 3

a) Consider a database of 2000 face  images of size 30 by 60. This database contains images of 40 people each having 50 images. Now consider applying Principal Component Analysis (PCA) to the database to construct 10 eigenfaces for face recognition.

i)   What is the dimensionality of the covariance matrix of the dataset?

ii)  What is the dimensionality of the mean face?

iii) What is the dimensionality of the eigenface?

iv) What is the dimensionality of the pattern vector?

[8 marks]

b) We have 12 images belonging to 4 people. We apply face recognition and get the results shown in the table below. Compute the recognition rate, and the confusion matrix.

Ground truth label

1

1

1

2

2

2

3

3

3

4

4

4

Predicted label

2

1

1

2

2

2

3

2

3

3

4

1

[5 marks]

c)  Suppose you are designing an eigenface based face recognition system for a company. The system is trained using the face images of all the employees of the company. When a human passes the entrance of the company building, a camera will capture the face of the human and perform. face recognition.

i) Which pre-processing technique is required before face recognition and why?

ii) How will you design the system to make it able to determine whether: the human is an employee of the company; the human is not an employee of the company; a non- face object is captured by the camera.

[12 marks]


Question 4

a) Explain why the Viola-Jones algorithm is slow in training but very fast in detection.

[6 marks]

b) Fig. 3(a) depicts a Haar feature pattern (white colour denotes +1 and black denotes -1) and Fig. 3(b) depicts the pixel values of a 4x4 image. Fig. 3(a) and Fig. 3(b) are of the same size. Compute the Haar feature with the following steps.

i) Compute the integral image of Fig. 3(b)

ii) Show the procedure to compute the Haar feature value from the integral image.

 

Fig. 3(a) Harr feature                            Fig. 3(b) Image

[10 marks]

c) Compare and contrast the information modelled in Active Shape Model (ASM) vs Active Appearance Model (AAM). Is the ASM suitable for modelling any deformable object or only faces? Explain you answers.

[9 marks]

 

 


热门主题

课程名

mktg2509 csci 2600 38170 lng302 csse3010 phas3226 77938 arch1162 engn4536/engn6536 acx5903 comp151101 phl245 cse12 comp9312 stat3016/6016 phas0038 comp2140 6qqmb312 xjco3011 rest0005 ematm0051 5qqmn219 lubs5062m eee8155 cege0100 eap033 artd1109 mat246 etc3430 ecmm462 mis102 inft6800 ddes9903 comp6521 comp9517 comp3331/9331 comp4337 comp6008 comp9414 bu.231.790.81 man00150m csb352h math1041 eengm4100 isys1002 08 6057cem mktg3504 mthm036 mtrx1701 mth3241 eeee3086 cmp-7038b cmp-7000a ints4010 econ2151 infs5710 fins5516 fin3309 fins5510 gsoe9340 math2007 math2036 soee5010 mark3088 infs3605 elec9714 comp2271 ma214 comp2211 infs3604 600426 sit254 acct3091 bbt405 msin0116 com107/com113 mark5826 sit120 comp9021 eco2101 eeen40700 cs253 ece3114 ecmm447 chns3000 math377 itd102 comp9444 comp(2041|9044) econ0060 econ7230 mgt001371 ecs-323 cs6250 mgdi60012 mdia2012 comm221001 comm5000 ma1008 engl642 econ241 com333 math367 mis201 nbs-7041x meek16104 econ2003 comm1190 mbas902 comp-1027 dpst1091 comp7315 eppd1033 m06 ee3025 msci231 bb113/bbs1063 fc709 comp3425 comp9417 econ42915 cb9101 math1102e chme0017 fc307 mkt60104 5522usst litr1-uc6201.200 ee1102 cosc2803 math39512 omp9727 int2067/int5051 bsb151 mgt253 fc021 babs2202 mis2002s phya21 18-213 cege0012 mdia1002 math38032 mech5125 07 cisc102 mgx3110 cs240 11175 fin3020s eco3420 ictten622 comp9727 cpt111 de114102d mgm320h5s bafi1019 math21112 efim20036 mn-3503 fins5568 110.807 bcpm000028 info6030 bma0092 bcpm0054 math20212 ce335 cs365 cenv6141 ftec5580 math2010 ec3450 comm1170 ecmt1010 csci-ua.0480-003 econ12-200 ib3960 ectb60h3f cs247—assignment tk3163 ics3u ib3j80 comp20008 comp9334 eppd1063 acct2343 cct109 isys1055/3412 math350-real math2014 eec180 stat141b econ2101 msinm014/msing014/msing014b fit2004 comp643 bu1002 cm2030
联系我们
EMail: 99515681@qq.com
QQ: 99515681
留学生作业帮-留学生的知心伴侣!
工作时间:08:00-21:00
python代写
微信客服:codinghelp
站长地图