1
00:00:06,139 --> 00:00:07,540
Hello, everyone.

2
00:00:07,540 --> 00:00:10,810
Welcome to the next episode of Hewlett
Packard Lab podcast from research

3
00:00:10,810 --> 00:00:12,078
to reality.

4
00:00:12,078 --> 00:00:15,815
This time, we are doing
the spotlight on animal tracking.

5
00:00:15,815 --> 00:00:18,818
I have amazing pleasure and honor

6
00:00:18,818 --> 00:00:22,722
to host two gentlemen here in the room
and one gentleman out of Africa.

7
00:00:23,189 --> 00:00:24,891
let's start with a remote.

8
00:00:24,891 --> 00:00:26,226
attendee.

9
00:00:26,226 --> 00:00:29,229
Alex, would you mind introducing yourself?

10
00:00:29,429 --> 00:00:31,264
Yes. Good evening to you.

11
00:00:31,264 --> 00:00:32,198
Thank you for having me.

12
00:00:32,198 --> 00:00:34,300
It's it's a great pleasure.

13
00:00:34,300 --> 00:00:37,904
I, I met, HPE,

14
00:00:37,904 --> 00:00:41,608
about two years ago
with the idea of creating this app

15
00:00:42,008 --> 00:00:46,379
that would automatically be able
to identify animal tracks and, and prints

16
00:00:46,780 --> 00:00:48,081
and, I'm happy to say

17
00:00:48,081 --> 00:00:51,284
we are a long way down the track
and the signs are fantastic.

18
00:00:51,284 --> 00:00:52,752
And it looks like
we are getting this right.

19
00:00:54,354 --> 00:00:55,455
Thanks, Alex.

20
00:00:55,455 --> 00:00:58,591
I'm sure you'll
be able to tell us much more about it.

21
00:00:58,858 --> 00:01:01,261
Let's introduce, first. Shivang.

22
00:01:01,261 --> 00:01:02,495
Hi. Shivang. Hi Dejan.

23
00:01:02,495 --> 00:01:04,831
Nice to meet you again. yeah. Hi,
everyone.

24
00:01:04,831 --> 00:01:06,332
I'm Shivang and I.

25
00:01:06,332 --> 00:01:09,502
I'm a researcher at here
in the Networking and Distributed Systems

26
00:01:09,502 --> 00:01:12,739
Lab working
with, Puneet Sharma, our director.

27
00:01:12,939 --> 00:01:15,341
I've been here for about a year
and a half now.

28
00:01:15,341 --> 00:01:18,011
before that, I was finishing up my PhD

29
00:01:18,011 --> 00:01:21,514
at Northeastern University,
where I did my PhD in wireless networking.

30
00:01:21,848 --> 00:01:23,383
Thanks, Shivang! Hi Lianjie!

31
00:01:23,383 --> 00:01:25,051
Welcome back on the podcast.

32
00:01:25,051 --> 00:01:26,853
It's a pleasure to be back, Dejan.

33
00:01:26,853 --> 00:01:28,388
my name is Lianjie Cao.

34
00:01:28,388 --> 00:01:31,124
I'm a senior researcher at Hewlett Packard Labs.

35
00:01:31,124 --> 00:01:35,562
My journey with HP slash
HPE started with an internship

36
00:01:35,562 --> 00:01:38,998
back in 2013, and I joined Labs
as a full time researcher

37
00:01:38,998 --> 00:01:42,001
in 2018
after I got my PhD degree from Purdue.

38
00:01:42,168 --> 00:01:43,436
Okay, excellent.

39
00:01:43,436 --> 00:01:45,738
So I heard about this project
of animal tracking.

40
00:01:45,738 --> 00:01:48,708
I always wanted to learn more.
I never had a chance.

41
00:01:48,708 --> 00:01:52,145
So this is, you know, like,
a double opportunity both

42
00:01:52,145 --> 00:01:54,581
to talk to you guys
and to learn a little bit more.

43
00:01:54,581 --> 00:01:57,383
So who wants to, 
take the first stab at it?

44
00:01:57,383 --> 00:01:58,218
Yeah, I can start.

45
00:01:58,218 --> 00:02:01,488
so, yeah, this project is basically,

46
00:02:01,855 --> 00:02:04,858
quite an interesting project,
as you mentioned yourself.

47
00:02:05,325 --> 00:02:08,895
so at a very high level, it's
basically about trying to digitize

48
00:02:08,895 --> 00:02:11,998
and save this ancient art
that is called animal tracking.

49
00:02:12,365 --> 00:02:14,234
And and what is animal tracking?

50
00:02:14,234 --> 00:02:17,971
Basically, it's this ancient scale
that was evolved in Africa

51
00:02:17,971 --> 00:02:19,272
thousands of years ago.

52
00:02:19,272 --> 00:02:22,842
And basically what it involves
is these human trackers,

53
00:02:22,842 --> 00:02:27,280
they go out into these national parks,
forest wildlife reserves,

54
00:02:27,614 --> 00:02:33,286
they must form a mental image of what
the animals are doing,

55
00:02:33,286 --> 00:02:37,624
where they're going, and that they don't
do that by befalling the animals.

56
00:02:37,624 --> 00:02:40,860
They do that by looking at the signs
that the animals have left behind.

57
00:02:41,227 --> 00:02:43,196
from that could be their pawprints.

58
00:02:43,196 --> 00:02:46,199
That could be droppings
or other scratches that they leave behind.

59
00:02:46,599 --> 00:02:50,270
and so that's what, animal tracking is.

60
00:02:50,270 --> 00:02:52,038
And unfortunately, due to,

61
00:02:53,006 --> 00:02:54,541
urbanization that's happening globally.

62
00:02:54,541 --> 00:02:55,041
Right.

63
00:02:55,041 --> 00:02:58,111
this ancient art
is sort of withering away.

64
00:02:58,545 --> 00:03:02,582
And our goal here with Alex
is to try to save that ancient art.

65
00:03:02,582 --> 00:03:05,418
Right. And we are building an end
to end system for that.

66
00:03:05,418 --> 00:03:09,122
I'm sure we'll go into details, later,
but just briefly.

67
00:03:09,122 --> 00:03:10,156
We have a front end.

68
00:03:10,156 --> 00:03:13,726
We have a front end piece,
which is the mobile apps and the web apps

69
00:03:13,726 --> 00:03:17,864
and then which help us gather data
and get images from the field.

70
00:03:18,298 --> 00:03:20,967
And then we have our back end,
which has host and the intelligence.

71
00:03:20,967 --> 00:03:21,167
Right.

72
00:03:21,167 --> 00:03:25,572
The machine learning algorithms
and whatnot, to then take in

73
00:03:25,772 --> 00:03:29,709
as input those images and then output
which species they belong to and so on.

74
00:03:30,009 --> 00:03:32,445
So what was your first thought
when you heard about it?

75
00:03:32,445 --> 00:03:36,849
Lianjie, about Alex and his project,
which really is amazing.

76
00:03:37,183 --> 00:03:40,353
Shivang can start first because I
joined the project a little bit late.

77
00:03:40,353 --> 00:03:41,321
I can add to that.

78
00:03:41,321 --> 00:03:41,854
So yeah.

79
00:03:41,854 --> 00:03:46,793
So, yeah, for me,
I remember Puneet one day just came to me,

80
00:03:46,793 --> 00:03:51,097
and talked to me about this project,
and at first I was a little confused,

81
00:03:51,097 --> 00:03:54,100
and, like, I didn't know what to expect
from this project.

82
00:03:54,200 --> 00:03:59,439
but then, thinking in the whole sort of,
scope of the project,

83
00:03:59,439 --> 00:04:02,642
I was very excited, basically, and trying

84
00:04:02,642 --> 00:04:06,179
to see what all it entails.

85
00:04:06,179 --> 00:04:09,182
And it's, you know, very different
from typical research projects

86
00:04:09,482 --> 00:04:12,652
that at least I've taken or over
in my research career.

87
00:04:13,052 --> 00:04:15,655
And, and also

88
00:04:15,655 --> 00:04:20,159
when we we started speaking to Alex
for the first time,

89
00:04:20,159 --> 00:04:24,330
it really clearly came across
how passionate he is about the project.

90
00:04:24,330 --> 00:04:24,897
Right.

91
00:04:24,897 --> 00:04:27,767
And that itself is very motivating. Right?

92
00:04:27,767 --> 00:04:30,270
how passionate he is,
how inspiring he is.

93
00:04:30,270 --> 00:04:32,672
And that helps
us push ourselves further too.

94
00:04:33,840 --> 00:04:36,442
I mean, to
me, it's indeed a very different project,

95
00:04:36,442 --> 00:04:39,212
a different project,
when compared to the self

96
00:04:39,212 --> 00:04:42,215
entertaining kind of research project
that I do in labs.

97
00:04:42,482 --> 00:04:44,150
and I joined the project
a little bit late,

98
00:04:44,150 --> 00:04:46,452
mainly working on a machine
learning part of it.

99
00:04:46,452 --> 00:04:49,455
and, but the first time
I heard about the project, like,

100
00:04:49,656 --> 00:04:52,925
oh, man, finally we're shifting
our research directions

101
00:04:52,925 --> 00:04:56,663
a little bit and starting to work on
some, something fun.

102
00:04:56,796 --> 00:04:59,465
And I'm ready for an adventure
in the jungle.

103
00:04:59,465 --> 00:05:02,268
Right. So. But the thing is, like,
of course, we never do that.

104
00:05:02,268 --> 00:05:05,938
And, it's really a great and exciting,
project, like

105
00:05:06,673 --> 00:05:09,676
Shivang said, and a terrific experience
to work with experts

106
00:05:09,676 --> 00:05:14,047
like, like,
Alex from a completely different field.

107
00:05:14,280 --> 00:05:16,115
Okay,
so we've been talking about Alex this.

108
00:05:16,115 --> 00:05:17,517
Alex that. Alex.

109
00:05:17,517 --> 00:05:20,086
How did you approach HPE?

110
00:05:20,086 --> 00:05:23,156
Okay,
so I run an NGO, a not for profit company

111
00:05:23,156 --> 00:05:28,127
in South Africa called the Tracker
Academy, and we train young rural people

112
00:05:28,127 --> 00:05:32,131
from big wildlife areas
in traditional skills of animal tracking.

113
00:05:32,165 --> 00:05:35,034
A one-year formally accredited program,

114
00:05:35,034 --> 00:05:39,005
we are a school, and then we deploy
our graduates into conservation jobs.

115
00:05:39,539 --> 00:05:42,575
And it's just become evident
over the last few years

116
00:05:42,775 --> 00:05:47,013
how the ancient traditional skills
of tracking in Africa have diminished.

117
00:05:47,013 --> 00:05:49,449
We are losing these skills and waste.

118
00:05:49,449 --> 00:05:55,221
And at about the same time
I met with, Dobias van Ingen from HPE, he,

119
00:05:55,388 --> 00:05:59,025
from Aruba,
and he was in South Africa on a conference

120
00:05:59,359 --> 00:06:02,128
and we happened to meet
just coincidentally.

121
00:06:02,128 --> 00:06:05,565
And I told him about the idea
that we needed

122
00:06:05,565 --> 00:06:09,402
to digitize this ancient data
before we lose it completely.

123
00:06:09,702 --> 00:06:12,839
And and he took the idea back to HPE.

124
00:06:12,839 --> 00:06:14,574
And that was two years ago.

125
00:06:14,574 --> 00:06:18,378
And here we sit now with,
I think it's version three of the app.

126
00:06:18,678 --> 00:06:20,780
and we are a long way down the road.

127
00:06:20,780 --> 00:06:25,017
And I must say, there's been great success
and it's been such a pleasure to work

128
00:06:25,017 --> 00:06:29,288
with such an incredibly talented HPE team
and the HPE labs.

129
00:06:29,522 --> 00:06:32,425
What these people have been able to
achieve is just astonishing.

130
00:06:34,026 --> 00:06:34,861
So I can

131
00:06:34,861 --> 00:06:38,798
thank you Alex I can hear
all this excitement from your side,

132
00:06:38,798 --> 00:06:42,201
but I'm curious from far away Africa,

133
00:06:42,201 --> 00:06:45,271
how did it excite you here in Milpitas?

134
00:06:45,505 --> 00:06:46,906
Lianjie and Shivang. Sure.

135
00:06:46,906 --> 00:06:49,909
so I think to me it's the impact.

136
00:06:50,143 --> 00:06:53,479
So we have been working on
an innovative idea.

137
00:06:53,479 --> 00:06:57,450
like challenging problems
in the labs, but those ideas and those,

138
00:06:58,484 --> 00:07:00,820
challenging problems
usually take a very long time

139
00:07:00,820 --> 00:07:02,688
to see the impact, like, for years.

140
00:07:02,688 --> 00:07:04,957
But this project is very different
in the sense

141
00:07:04,957 --> 00:07:08,694
that we can directly, almost directly
see the impact immediately.

142
00:07:08,928 --> 00:07:12,932
And all the improvements and
all the enhancement we made in the system

143
00:07:13,166 --> 00:07:17,336
are going to be used by Alex and his team
on the field, like right away.

144
00:07:17,603 --> 00:07:19,705
So there's no way do you see the impact?

145
00:07:19,705 --> 00:07:24,010
And I think also the impact of sort of
like a beyond labs, even beyond HPE.

146
00:07:24,010 --> 00:07:25,411
Like we have been always saying

147
00:07:25,411 --> 00:07:28,648
accelerating the impact,
being a force for good.

148
00:07:29,282 --> 00:07:29,615
Yeah.

149
00:07:29,615 --> 00:07:31,984
And I think, to add to what
Lianjie said,

150
00:07:31,984 --> 00:07:34,353
he mentioned a bit about the social 
impact.

151
00:07:34,353 --> 00:07:37,824
But, I mean, even from the technical side,
as researchers,

152
00:07:37,824 --> 00:07:40,860
right, we get excited about tackling
challenging problems.

153
00:07:40,860 --> 00:07:43,796
And this, again, I did not
you know, we have

154
00:07:43,796 --> 00:07:47,333
we should expect challenging these sort
of challenging problems at the beginning.

155
00:07:47,567 --> 00:07:50,636
But as we went through the project,
we encountered several challenges

156
00:07:50,636 --> 00:07:53,072
and technical challenges
that we had to overcome.

157
00:07:53,072 --> 00:07:55,341
And as researchers,
nothing more exciting than that, right?

158
00:07:55,341 --> 00:07:55,975
Yeah.

159
00:07:55,975 --> 00:08:00,413
So I can see why historically
this skill is very important.

160
00:08:00,680 --> 00:08:03,683
But how is it still important today? Alex.

161
00:08:04,150 --> 00:08:06,118
Yes, that's a good question.

162
00:08:06,118 --> 00:08:09,856
Today, in modern day conservation efforts,

163
00:08:10,089 --> 00:08:15,394
we need trackers to track to collect data
on behavior of animals

164
00:08:15,862 --> 00:08:18,865
and the trackers,
but possess the knowledge

165
00:08:19,198 --> 00:08:22,235
able to to accurately identify
and interpret

166
00:08:22,568 --> 00:08:26,138
the signs of the animals left by animals
and thereby able

167
00:08:26,138 --> 00:08:30,076
to collect important data that otherwise
would not have been collected.

168
00:08:30,977 --> 00:08:34,247
that trackers are able to know
what animals are doing

169
00:08:34,514 --> 00:08:38,784
even when they're not there, by
by following and interpreting their signs.

170
00:08:39,085 --> 00:08:41,454
We also need trackers in anti-poaching.

171
00:08:41,454 --> 00:08:42,688
I'm not sure if you know,

172
00:08:42,688 --> 00:08:46,325
but there's been a crisis
in southern Africa of rhino poaching.

173
00:08:46,526 --> 00:08:49,962
We've lost something
like 70% of our rhino population

174
00:08:50,196 --> 00:08:54,767
in southern Africa over the last decade
and a half, and we need talented

175
00:08:54,767 --> 00:08:59,105
trackers who can pursue the poachers
on foot and track and find them.

176
00:08:59,505 --> 00:09:03,109
So it's, 
the trackers are the unsung heroes,

177
00:09:03,109 --> 00:09:07,246
and their skills are becoming
increasingly important

178
00:09:07,246 --> 00:09:11,117
as as we go through time
and as conservation management

179
00:09:11,217 --> 00:09:14,420
start to realize the value,
the tangible value

180
00:09:14,420 --> 00:09:17,823
that this ancient skill brings,
the modern day conservation efforts.

181
00:09:20,426 --> 00:09:21,661
So I can understand

182
00:09:21,661 --> 00:09:25,464
people in the field,
and all the challenges they have.

183
00:09:25,464 --> 00:09:26,966
I can imagine them.

184
00:09:26,966 --> 00:09:30,536
But what are the challenges for
you here in Milpitas?

185
00:09:30,603 --> 00:09:34,974
Shivang and Lianjie,
I can start, like.

186
00:09:34,974 --> 00:09:38,411
So our goal, the HPE team is to build

187
00:09:38,411 --> 00:09:42,348
a digitized solution
for Alex to solve all the challenges.

188
00:09:42,982 --> 00:09:44,016
he mentioned.

189
00:09:44,016 --> 00:09:47,186
And the the challenges for building such
systems is, like, every step of the way.

190
00:09:47,420 --> 00:09:49,655
So if you think of this,
like starting with,

191
00:09:49,655 --> 00:09:53,059
we need to figure out the design
of the architecture

192
00:09:53,292 --> 00:09:58,030
for this end to end system,
from capturing the footprint in the jungle

193
00:09:58,030 --> 00:10:02,301
to infer the images of the species, 
in the back end.

194
00:10:02,468 --> 00:10:02,668
Right.

195
00:10:02,668 --> 00:10:07,907
So, for instance, we need a front end
to capture the footprints and annotate

196
00:10:08,007 --> 00:10:11,510
the footprint with the information
related information before they get lost.

197
00:10:11,744 --> 00:10:16,082
And we also need to upload them
reliably, to the back end.

198
00:10:16,082 --> 00:10:19,285
So all those kind of things
we need to take care of even in the back end,

199
00:10:19,285 --> 00:10:22,254
we also need to think about
how should we clean the image,

200
00:10:22,254 --> 00:10:26,626
how should we store them, label them
and then infer the

201
00:10:26,626 --> 00:10:29,095
the species of the footprint,
all those kind of stuff?

202
00:10:30,329 --> 00:10:30,596
Yeah.

203
00:10:30,596 --> 00:10:34,367
And to add to
that, I mean, we have way more challenges

204
00:10:34,367 --> 00:10:38,471
because of the fact that
the geography of the, the problem itself.

205
00:10:38,471 --> 00:10:39,005
Right.

206
00:10:39,005 --> 00:10:41,974
A lot of these trackers are working
in remote areas where there is like

207
00:10:41,974 --> 00:10:44,110
little to no connectivity at all. Right.

208
00:10:44,110 --> 00:10:47,346
And we are asking them
to upload these images from those areas.

209
00:10:47,346 --> 00:10:48,681
So how do they do that?

210
00:10:48,681 --> 00:10:49,281
There are all sorts

211
00:10:49,281 --> 00:10:52,785
of technical challenges that arise
because of that lack of connectivity.

212
00:10:53,686 --> 00:10:55,821
Are there any non-technical challenges?

213
00:10:55,821 --> 00:10:56,188
Alex.

214
00:10:56,188 --> 00:10:58,457
Because here we only jump on technology.

215
00:10:58,457 --> 00:10:59,425
Technology.

216
00:10:59,425 --> 00:11:02,261
I'm sure there are other kinds of problems
that you are,

217
00:11:02,261 --> 00:11:05,097
exposed to on a daily basis.

218
00:11:05,097 --> 00:11:05,631
Well, yes.

219
00:11:05,631 --> 00:11:11,170
In in training the machine, the algorithm,
we need to collect many, many examples.

220
00:11:11,170 --> 00:11:15,641
Let's take for example, we want to train
the machine to identify a lions track.

221
00:11:15,941 --> 00:11:19,645
We need to to, to up to identify,

222
00:11:19,912 --> 00:11:24,884
photograph and upload at least a thousand
different examples of lion tracks.

223
00:11:25,151 --> 00:11:26,619
And so that takes time.

224
00:11:26,619 --> 00:11:32,224
And you so and it takes people
that who know how to identify

225
00:11:32,224 --> 00:11:33,559
a lion track easily.

226
00:11:33,559 --> 00:11:37,129
If we have less than competent trackers
training the machine,

227
00:11:37,496 --> 00:11:40,232
it's, it's going to have a great impact
on the on

228
00:11:40,232 --> 00:11:43,235
the outcome of the,
the the performance of of the app.

229
00:11:43,335 --> 00:11:47,707
So we've got to make sure we've got expert
trackers out in the field collecting

230
00:11:47,707 --> 00:11:53,412
literally thousands and thousands
of images of 120 different species.

231
00:11:53,679 --> 00:11:55,748
So that's an incredible feat.

232
00:11:55,748 --> 00:11:58,484
And it's it takes time
and weather often is a problem.

233
00:11:58,484 --> 00:12:02,421
so yes, that's that would be the most,

234
00:12:02,421 --> 00:12:06,292
apparent challenge we are dealing with,
but we're getting through it slowly.

235
00:12:06,592 --> 00:12:10,129
I think we've, collected
about 30,000 images so far.

236
00:12:11,797 --> 00:12:13,399
So you gave me

237
00:12:13,399 --> 00:12:17,336
some hint that you are using an app
to solve the problems.

238
00:12:17,336 --> 00:12:20,573
Can you tell me how you are
solving the problems using this app?

239
00:12:21,941 --> 00:12:22,875
sure.

240
00:12:22,875 --> 00:12:27,012
So, for the app side, we basically have
like two different types of apps.

241
00:12:27,012 --> 00:12:30,783
We have the mobile app and installed
on the smartphone of the trackers.

242
00:12:31,016 --> 00:12:33,219
And we also have the web app.

243
00:12:33,219 --> 00:12:37,256
So for the mobile app, 
the users can create

244
00:12:37,523 --> 00:12:41,227
different type of account like public account 
and professional account and things like that.

245
00:12:41,460 --> 00:12:45,865
And then in their account they can browse
first all the images you have taken

246
00:12:45,865 --> 00:12:48,934
before with the different types
of information embedded.

247
00:12:49,235 --> 00:12:52,705
And they can also, of course
take new photos of the footprints

248
00:12:52,972 --> 00:12:56,275
and then at the same time
associate the metadata information

249
00:12:56,275 --> 00:12:59,245
of the of the species
and things like that.

250
00:12:59,411 --> 00:13:03,716
And then and finally they can upload
your images to the back end.

251
00:13:04,016 --> 00:13:07,586
And in the web app, the app,
the web app is totally different.

252
00:13:07,753 --> 00:13:11,757
It's more for the purpose
of analysis and management.

253
00:13:11,991 --> 00:13:14,994
So in the web apps,
the administrator can go ahead

254
00:13:14,994 --> 00:13:19,165
and browse all the images uploaded
by different users.

255
00:13:19,231 --> 00:13:21,801
Professional, public
and all those kind of stuff.

256
00:13:21,801 --> 00:13:25,237
And they can check
into the detailed information of them

257
00:13:25,237 --> 00:13:28,674
and try to, modify
the information of some of them are wrong.

258
00:13:28,874 --> 00:13:33,045
And what's more interesting is
they can also browse the images,

259
00:13:33,145 --> 00:13:38,717
filter them and show them or, you know,
the active map to see the geographical,

260
00:13:38,717 --> 00:13:40,352
you know, information on the map

261
00:13:40,352 --> 00:13:43,022
so they can get a better visual
understanding

262
00:13:43,022 --> 00:13:46,025
of where the animals are spotted
and all this kind of stuff.

263
00:13:46,292 --> 00:13:50,095
And in the future,
we are also thinking of doing more deep,

264
00:13:50,095 --> 00:13:52,164
deeper kind of like analysis

265
00:13:52,164 --> 00:13:55,901
and understanding based on this type
of geographical information.

266
00:13:56,068 --> 00:14:00,306
For instance, we can try to learn
how global warming impacts

267
00:14:00,306 --> 00:14:04,143
the animal migration at a very high level
and things like that,

268
00:14:04,610 --> 00:14:08,113
but also like to highlight the fact that,
I mean, the challenges that we describe

269
00:14:08,113 --> 00:14:10,216
this, such a diverse set of challenges.

270
00:14:10,216 --> 00:14:13,285
And so we needed a diverse set of people
with diverse skill sets

271
00:14:13,786 --> 00:14:14,887
to tackle those problems. Right.

272
00:14:14,887 --> 00:14:17,656
So this effort is sort of a pan-HPE
effort.

273
00:14:17,656 --> 00:14:21,594
Over the time we've gotten more
and more collaborators from outside labs.

274
00:14:21,727 --> 00:14:25,798
So for example,
we have Ezmeral BU

275
00:14:25,798 --> 00:14:29,168
who are providing us with the back end
to help us run the modeling

276
00:14:29,168 --> 00:14:34,707
and the inference on, then we have,
the Asia Pacific Innovation Center team.

277
00:14:34,974 --> 00:14:35,641
Who is

278
00:14:35,641 --> 00:14:37,610
Who's helping us
build the apps themselves. Right.

279
00:14:37,610 --> 00:14:40,045
Because, I mean,
these apps need to be very professional.

280
00:14:40,045 --> 00:14:41,046
Grade.

281
00:14:41,046 --> 00:14:43,515
We can build some apps,
but I'm sure they won't be

282
00:14:43,515 --> 00:14:45,150
as good as what these guys put out.

283
00:14:45,150 --> 00:14:47,586
I mean, these guys are really talented
at what they do.

284
00:14:47,586 --> 00:14:50,022
so yeah, that's
just something I want to highlight

285
00:14:50,022 --> 00:14:54,560
I'm glad you started to talk about business.

286
00:14:55,594 --> 00:14:57,596
you have so far

287
00:14:57,596 --> 00:15:02,568
received help from various businesses
like Ezmeral or Asia-Pacific teams.

288
00:15:03,068 --> 00:15:07,006
But in the big picture,
how do you see this becoming sustainable?

289
00:15:07,306 --> 00:15:10,075
Not necessarily making money out of it,
but someone needs,

290
00:15:10,075 --> 00:15:13,345
on a daily basis, annual basis
to support this.

291
00:15:13,345 --> 00:15:17,116
So it should eventually go
into some business, into some production.

292
00:15:17,449 --> 00:15:20,419
How do you gentlemen see this?

293
00:15:20,419 --> 00:15:24,356
so I think, the idea who we are

294
00:15:24,356 --> 00:15:27,393
at the moment, currently we are relying
more on the goodwill of, let's say,

295
00:15:27,393 --> 00:15:31,897
HPE and Alex that, we can sustain
this project, in the short term.

296
00:15:32,064 --> 00:15:35,634
But I think in the long term,
it can also be about once we

297
00:15:35,634 --> 00:15:39,104
if you want to monetarily sustain
it can also be about

298
00:15:39,338 --> 00:15:43,609
how we gamify the app a little bit,
maybe for like folks like you and me.

299
00:15:43,609 --> 00:15:43,809
Right.

300
00:15:43,809 --> 00:15:46,946
Like we can go out in the field, snap,

301
00:15:46,946 --> 00:15:51,116
you have pictures of footprints
and send it up and get a result.

302
00:15:51,116 --> 00:15:53,986
Right? Okay. This is a lion’s footprint.
This is a wolf’s footprint.

303
00:15:53,986 --> 00:15:57,022
And then maybe that builds
this sort of ecosystem of users

304
00:15:57,022 --> 00:16:00,125
that, the an end goal could be.

305
00:16:00,125 --> 00:16:03,662
And then that could ultimately help us
sustain, what we're doing.

306
00:16:03,662 --> 00:16:07,700
And hopefully, I mean,
we always will rely as, again, HPE's

307
00:16:07,733 --> 00:16:09,435
nature of being a force for good, right?

308
00:16:09,435 --> 00:16:12,338
I mean, that's always a thing
that will hopefully help us

309
00:16:12,338 --> 00:16:13,906
sustain this in the long term. Yeah.

310
00:16:13,906 --> 00:16:15,374
And to add a little bit to that.

311
00:16:15,374 --> 00:16:20,145
So this is becoming something that the
the impact of this project is going beyond

312
00:16:20,145 --> 00:16:25,217
HPE because we have been talking to
and Alex as well, talking to some other

313
00:16:25,217 --> 00:16:27,720
service providers,
they are willing to provide

314
00:16:27,720 --> 00:16:31,457
like hosting services to the project
because you want to advertise, you know,

315
00:16:31,490 --> 00:16:33,959
of course,
they're on those and things like that.

316
00:16:33,959 --> 00:16:37,262
So those can become a more,
you know, general coverage team

317
00:16:37,262 --> 00:16:39,031
across companies and things like that

318
00:16:39,031 --> 00:16:42,234
and be a very good showcase of HPE's
technology.

319
00:16:42,234 --> 00:16:45,304
It can be, 
can be applied to solve such problems.

320
00:16:45,304 --> 00:16:46,205
So we can get some,

321
00:16:46,205 --> 00:16:50,776
you know, some similar attentions
and projects from potential customers.

322
00:16:51,243 --> 00:16:55,481
Any opportunity for our Aruba
because they are our edge side.

323
00:16:55,948 --> 00:16:56,281
Yeah.

324
00:16:56,281 --> 00:16:58,684
So yeah, as I mentioned briefly before.
Right.

325
00:16:58,684 --> 00:17:02,121
the work that a lot of
these trackers are doing are in

326
00:17:02,187 --> 00:17:05,391
remote areas where there is
basically no connectivity at all.

327
00:17:05,391 --> 00:17:06,058
Right.

328
00:17:06,058 --> 00:17:09,261
so Aruba, 
basically us working together with Aruba

329
00:17:09,261 --> 00:17:12,865
can help us tackle
that specific part of this problem where,

330
00:17:13,265 --> 00:17:17,569
maybe using technologies
like satellite or private 5G,

331
00:17:17,836 --> 00:17:20,839
we can have some deployments
in these remote areas.

332
00:17:21,206 --> 00:17:24,810
that can help get connectivity
to those folks that don't have it.

333
00:17:24,810 --> 00:17:25,344
Right.

334
00:17:25,344 --> 00:17:30,349
And that will ultimately help us reduce
the latency that these guys face in terms

335
00:17:30,349 --> 00:17:34,286
of uploading the images, getting results
back from the servers and so on.

336
00:17:34,286 --> 00:17:37,756
And if we can do that, obviously
that will ultimately help the quality

337
00:17:37,756 --> 00:17:41,693
of experience that these guys, face
in these remote areas.

338
00:17:42,361 --> 00:17:45,697
So accomplishing what you have,
which is really outstanding,

339
00:17:45,764 --> 00:17:49,034
what was the most challenging about it,
and was there anything

340
00:17:49,034 --> 00:17:52,538
unusual in terms of your solutions
or surprising things?

341
00:17:53,105 --> 00:17:57,142
I think to me the most challenging part
is actually labeling the images,

342
00:17:57,476 --> 00:18:00,946
which is kind of like strange
because Alex came to us for help,

343
00:18:01,513 --> 00:18:05,818
and it turns out that we need his help
to label the images first

344
00:18:06,085 --> 00:18:09,655
before we can apply technologies
like cloud computing

345
00:18:09,655 --> 00:18:11,056
and machine learning to help him.

346
00:18:11,056 --> 00:18:13,559
So it's it becomes some sort of
like a chicken egg problem.

347
00:18:13,559 --> 00:18:16,695
So. Well, you want me to help you,
then you need to help me first.

348
00:18:16,695 --> 00:18:19,398
That kind of stuff.
Which is kind of surprising to me.

349
00:18:19,398 --> 00:18:23,702
and if you think about this, like, what's
worst in this case, in this problem, is

350
00:18:24,803 --> 00:18:26,138
we there is nobody

351
00:18:26,138 --> 00:18:31,310
else except Alex and his team
can help us to solve this problem.

352
00:18:31,310 --> 00:18:33,312
Because these guys are professional for me.

353
00:18:33,312 --> 00:18:34,079
Like, I can tell

354
00:18:34,079 --> 00:18:37,182
the difference between cats and dogs,
but I cannot tell the difference

355
00:18:37,182 --> 00:18:38,183
from the footprint. Right.

356
00:18:38,183 --> 00:18:42,988
So and this is also an in from the other,
perspective, it's like,

357
00:18:43,455 --> 00:18:47,259
there are always something we take for
granted in our research project,

358
00:18:47,259 --> 00:18:50,963
but in reality, those kind of stuff
like a label that you measure

359
00:18:51,029 --> 00:18:54,032
and things like that
are probably not quite there

360
00:18:54,399 --> 00:18:55,667
And, Alex.

361
00:18:55,667 --> 00:18:58,704
Have you considered any standardization?

362
00:18:59,071 --> 00:19:02,875
I mean, there are different
classes of animal tracks and all of that.

363
00:19:04,543 --> 00:19:09,481
Yes. Well, there's standardization
in terms of trackers skill set.

364
00:19:09,781 --> 00:19:13,819
we are able to objectively evaluate
a tracker’s

365
00:19:13,819 --> 00:19:19,291
skill set in the various components,
practical components of tracking.

366
00:19:19,525 --> 00:19:20,826
And that is what is established.

367
00:19:20,826 --> 00:19:23,829
And, 
the South African Department of Education

368
00:19:23,829 --> 00:19:27,132
recognizes, our standards in that regard.

369
00:19:27,266 --> 00:19:30,202
And there are standards
in the US as well, run

370
00:19:30,202 --> 00:19:33,605
by other companies that that promote
and do tracking assessments.

371
00:19:34,139 --> 00:19:37,943
But in terms of, standardizing the data,
that's very hard.

372
00:19:37,943 --> 00:19:42,181
And that's I just wanted to add
the most complex part of this program,

373
00:19:42,247 --> 00:19:46,919
this whole project
is that a lion footprint.

374
00:19:47,319 --> 00:19:50,322
No two lions footprints are the same.

375
00:19:50,589 --> 00:19:53,659
And it depends on the type of soil
that it stands in

376
00:19:54,159 --> 00:19:56,929
or how it's moving,
if it's turning, if it's running,

377
00:19:56,929 --> 00:20:00,732
if it's old or young, it's
constantly throwing up different types

378
00:20:00,732 --> 00:20:04,937
of, of of forms
of, of the same species track.

379
00:20:05,170 --> 00:20:08,507
And that's why we have to, 
get, get so many,

380
00:20:08,540 --> 00:20:12,377
capture so much data
to be able to train, train the machine.

381
00:20:12,411 --> 00:20:15,681
So that is, that is,
that is the most difficult part of it.

382
00:20:15,681 --> 00:20:20,919
And you know, we there there are similar
apps out there that identify, identify

383
00:20:20,919 --> 00:20:24,790
plants and trees by taking a photograph
of the leaf or the flower.

384
00:20:25,190 --> 00:20:29,828
That's a that's a far
simpler, procedure and project

385
00:20:29,828 --> 00:20:35,200
because there's, there's much less
very variation between an apples

386
00:20:35,200 --> 00:20:39,471
leaf that grows in the north of the US
and then in Europe,

387
00:20:39,771 --> 00:20:44,343
whereas in, in, within tracks, as I said,
there's just a great variability.

388
00:20:46,311 --> 00:20:47,546
And it

389
00:20:47,546 --> 00:20:50,816
seems intuitive to me that this is
very ethical, what you are doing.

390
00:20:50,816 --> 00:20:54,886
But have you ever considered
that angle of ethics, of animal tracking?

391
00:20:55,988 --> 00:20:56,321
Yeah.

392
00:20:56,321 --> 00:21:01,126
I mean, at the Tracker Academy
we have a whole module on ethics

393
00:21:01,426 --> 00:21:05,230
because many of our graduates
go into protected

394
00:21:05,230 --> 00:21:09,868
areas, government protected areas,
and have to protect endangered species.

395
00:21:10,235 --> 00:21:10,636
Excuse me.

396
00:21:10,636 --> 00:21:14,840
And so they have to
they have to undergo polygraph

397
00:21:14,840 --> 00:21:17,843
testing every, every six months.

398
00:21:18,210 --> 00:21:21,213
and their livelihood
depends on being ethical.

399
00:21:21,480 --> 00:21:25,317
and also from from another
completely different standpoint,

400
00:21:25,951 --> 00:21:30,088
the, the,
the observation of an animal's track

401
00:21:30,088 --> 00:21:34,359
or sign
and thereby being able to interpret

402
00:21:34,359 --> 00:21:39,865
what the animal is doing
is a completely noninvasive, approach.

403
00:21:40,232 --> 00:21:43,235
You don't have to go off to the animal
and dot it

404
00:21:43,402 --> 00:21:47,572
and put a collar around it and caught it
and give it drugs.

405
00:21:47,873 --> 00:21:50,676
Trackers are able to tell
where animals are moving

406
00:21:50,676 --> 00:21:53,979
and what they're doing, as I said earlier,
without even having to see them.

407
00:21:54,379 --> 00:21:59,685
And and that's very much goes
to this idea of sustainability,

408
00:21:59,685 --> 00:22:02,788
of low impact on the animals,
not disturbing them.

409
00:22:03,055 --> 00:22:06,058
And I think that that certainly falls
under the, under

410
00:22:06,058 --> 00:22:09,194
the umbrella of, of ethical conduct.

411
00:22:10,996 --> 00:22:11,430
Thank you.

412
00:22:11,430 --> 00:22:11,897
Alex.

413
00:22:11,897 --> 00:22:16,068
Coming back to Milpitas from Africa,
what did you two gentlemen find

414
00:22:16,068 --> 00:22:20,072
most rewarding in pursuing this project,
other than being extremely exciting?

415
00:22:21,206 --> 00:22:22,374
Yeah, I can start. yep.

416
00:22:22,374 --> 00:22:25,444
So for me,
I think, as you mentioned a few times

417
00:22:25,444 --> 00:22:29,448
already, it's just the direct impact
that is there of the project.

418
00:22:29,448 --> 00:22:31,950
I mean, all the projects we do,
I mean, we would like to think

419
00:22:31,950 --> 00:22:35,387
that they ultimately help society
in some way or the other down the line.

420
00:22:35,387 --> 00:22:35,754
Right.

421
00:22:35,754 --> 00:22:40,525
But with this project, you can really see
the directness of the impact.

422
00:22:40,525 --> 00:22:44,563
Like whatever we do,
it goes into the hands of these trackers

423
00:22:44,563 --> 00:22:47,566
that are going out into the wild.

424
00:22:47,699 --> 00:22:50,335
and ultimately been the goal, obviously,

425
00:22:50,335 --> 00:22:53,705
is to contribute
to the well-being, of our planet.

426
00:22:53,705 --> 00:22:56,575
Right. Finally.
That's what, we care about.

427
00:22:56,575 --> 00:22:59,578
So that's really the most rewarding part
for me.

428
00:22:59,945 --> 00:23:03,849
Yeah, to me, like,
I learned how to track animals, of course.

429
00:23:04,216 --> 00:23:06,918
And the other thing
I also realized through the project,

430
00:23:06,918 --> 00:23:09,955
it's basically the gap between research
project and real world

431
00:23:11,022 --> 00:23:12,023
problems.

432
00:23:12,023 --> 00:23:13,525
For instance, the problems

433
00:23:13,525 --> 00:23:17,763
that Shivang said, some of them are probably not
technically difficult to solve,

434
00:23:17,929 --> 00:23:22,234
but we do need to, to think from user's
perspective wearing their shoes to

435
00:23:22,234 --> 00:23:26,104
identify and to realize the problem before
we can do something about it, though.

436
00:23:26,171 --> 00:23:29,074
That's the thing I learned. And, Alex.

437
00:23:29,074 --> 00:23:33,445
And, Alex, why, does the audience
why does the public care about this?

438
00:23:34,279 --> 00:23:35,614
Now? That's a good question.

439
00:23:35,614 --> 00:23:38,350
The public should care about it
for a few reasons.

440
00:23:38,350 --> 00:23:41,753
One, it is part of ancient
African cultural heritage.

441
00:23:42,320 --> 00:23:45,157
and it has it

442
00:23:45,157 --> 00:23:47,492
certainly in South Africa, due

443
00:23:47,492 --> 00:23:50,495
to the effects of the apartheid regime

444
00:23:50,962 --> 00:23:54,866
and, and forcing people off their land,

445
00:23:55,133 --> 00:23:58,570
as well as the
the rapid worldwide trend of urbanization,

446
00:23:59,271 --> 00:24:02,707
people have forgotten
how to and lost the skills of tracking.

447
00:24:03,308 --> 00:24:08,480
And what I have seen
is that when we engage rural people,

448
00:24:08,513 --> 00:24:13,585
many of whom are unemployed,
living on the outskirts of these big

449
00:24:13,585 --> 00:24:16,588
wildlife areas, some of the last

450
00:24:16,688 --> 00:24:19,691
viable wildlife areas left on the planet,

451
00:24:20,192 --> 00:24:23,628
and they stare jealously
through the high wire fences.

452
00:24:24,262 --> 00:24:28,066
One way to capture the hearts
and minds of these people

453
00:24:28,433 --> 00:24:32,571
is through their own traditions, that
being the traditional skills of tracking.

454
00:24:32,938 --> 00:24:35,941
And so it has a twofold function for me.

455
00:24:36,074 --> 00:24:39,277
it's it's a skill
that can be used in a, in a technical

456
00:24:39,277 --> 00:24:42,280
manner
to better protect, endangered species.

457
00:24:42,347 --> 00:24:46,218
But it's also a skill that goes directly
to the hearts and minds of the people

458
00:24:46,485 --> 00:24:50,188
who are the custodians of these wild areas
and these,

459
00:24:50,655 --> 00:24:52,224
and these endangered species.

460
00:24:52,224 --> 00:24:55,894
So that's why I think the public
should be caring about this project.

461
00:24:57,529 --> 00:24:58,930
Great explanation.

462
00:24:58,930 --> 00:25:02,400
So we came close to the end,
of this discussion.

463
00:25:02,634 --> 00:25:06,271
Can we just close by,
each one of you very briefly stating,

464
00:25:06,271 --> 00:25:09,274
what do you do
when you don't track animals?

465
00:25:09,608 --> 00:25:14,579
Yeah, I can start like, I basically do
hikes, like a long hikes like five miles.

466
00:25:14,579 --> 00:25:15,413
Ten miles.

467
00:25:15,413 --> 00:25:17,115
I play badminton from time to time.

468
00:25:17,115 --> 00:25:18,683
Used to play more often, but much less.

469
00:25:18,683 --> 00:25:23,488
Now I play soccer sometimes I
also do a little bit woodworking at home.

470
00:25:23,488 --> 00:25:26,057
I have a small garage shop kind of stuff.

471
00:25:26,057 --> 00:25:26,925
Nice. Yeah.

472
00:25:26,925 --> 00:25:28,560
Thank you. Yeah. For me, I'm.

473
00:25:28,560 --> 00:25:32,597
I'm getting a big sports guy,
so I like to play a lot of sports, like,

474
00:25:32,697 --> 00:25:34,966
tennis, badminton, cricket.

475
00:25:34,966 --> 00:25:38,003
I don't know if people know about,
but it's a great sport.

476
00:25:38,870 --> 00:25:43,041
it, and, I seen with,
watching a lot of these sports as well.

477
00:25:43,041 --> 00:25:47,145
Then I like to attend, concerts
of artists, music artists that I like,

478
00:25:47,412 --> 00:25:48,747
like watching stand up comedy.

479
00:25:48,747 --> 00:25:49,881
That's authentic.

480
00:25:49,881 --> 00:25:51,950
Thanks. What about you, Alex?

481
00:25:51,950 --> 00:25:55,220
I love I love Shivang's answer. 

482
00:25:55,554 --> 00:25:58,890
I have two daughters
they are six years old and eight years old.

483
00:25:58,890 --> 00:26:03,061
They were both born at Londolozi Game
Reserve, where I worked for 23 years.

484
00:26:03,862 --> 00:26:07,265
we've now moved to the town
because of, their schooling.

485
00:26:07,265 --> 00:26:08,867
We needed a better school for them.

486
00:26:08,867 --> 00:26:12,504
But what I do do a lot of is
take them into wild areas,

487
00:26:12,771 --> 00:26:14,272
and I'm trying to introduce them.

488
00:26:14,272 --> 00:26:18,209
Well, I'm not trying, but I'm introducing
them to to tracking and tracking.

489
00:26:18,209 --> 00:26:22,213
It's just such a wonderful way
to get people to immerse themselves

490
00:26:22,213 --> 00:26:26,284
in wildlife,
to become connected with nature.

491
00:26:26,651 --> 00:26:30,155
And I'm seeing the vehicle of tracking,
doing just that.

492
00:26:30,455 --> 00:26:33,692
And I love spending time in nature
with my two little daughters.

493
00:26:35,126 --> 00:26:36,328
Thank you very much, Alex.

494
00:26:36,328 --> 00:26:40,231
Virtual handshake to you
and handshake here to Shivang and Lianjie.

495
00:26:40,231 --> 00:26:41,700
I really enjoyed this.

496
00:26:41,700 --> 00:26:43,234
I'm sure our audience will too.