WEBVTT

00:00:01.008 --> 00:00:04.489
<v Speaker 0>Our first finalist is Talon Rimmer from the Department of

NOTE CONF {"raw":[99,99,99,99,0,96,99,99,99,99]}

00:00:04.500 --> 00:00:05.260
<v Speaker 0>Biology.

NOTE CONF {"raw":[99]}

00:00:05.369 --> 00:00:08.789
<v Speaker 0>Talon's presentation is titled Farms Fish and Robots.

NOTE CONF {"raw":[5,99,99,99,99,99,99,99]}

00:00:10.520 --> 00:00:12.489
<v Speaker 1>Imagine you've achieved the dream.

NOTE CONF {"raw":[99,99,99,99,99]}

00:00:13.119 --> 00:00:16.149
<v Speaker 1>You've quit your job or thesis and started a new

NOTE CONF {"raw":[98,99,99,99,81,99,99,99,99,99]}

00:00:16.159 --> 00:00:20.680
<v Speaker 1>life as a full time commercial gardener and your friends

NOTE CONF {"raw":[99,99,99,99,88,99,99,99,99,99]}

00:00:20.690 --> 00:00:23.350
<v Speaker 1>have told you they will support you and buy whatever

NOTE CONF {"raw":[98,99,99,99,99,99,99,99,99,99]}

00:00:23.360 --> 00:00:27.409
<v Speaker 1>you grow provided you can prove that your garden also

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:00:27.420 --> 00:00:30.670
<v Speaker 1>serves as habitat for animals that live in the area.

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:00:31.450 --> 00:00:35.310
<v Speaker 1>The question is, how could you prove that your farm

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:00:35.319 --> 00:00:38.150
<v Speaker 1>does act as habitat for creatures in the long term?

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,98]}

00:00:38.970 --> 00:00:41.569
<v Speaker 1>Well, there are companies faced with the same question in

NOTE CONF {"raw":[99,99,99,99,89,78,99,99,99,99]}

00:00:41.580 --> 00:00:44.950
<v Speaker 1>British Columbia right now, but rather than growing things on

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:00:44.959 --> 00:00:47.549
<v Speaker 1>land, they grow their crop in the ocean.

NOTE CONF {"raw":[99,99,99,99,99,99,99,99]}

00:00:48.150 --> 00:00:51.560
<v Speaker 1>Seaweed farming is the practice of growing seaweed, a plant

NOTE CONF {"raw":[99,99,99,93,99,99,99,99,99,99]}

00:00:51.569 --> 00:00:54.610
<v Speaker 1>like organism similar to how you might grow a garden

NOTE CONF {"raw":[69,99,80,99,99,99,99,99,99,99]}

00:00:54.619 --> 00:00:55.290
<v Speaker 1>on land.

NOTE CONF {"raw":[99,99]}

00:00:55.409 --> 00:00:58.840
<v Speaker 1>But instead rows of these plants are attached to ropes

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:00:58.849 --> 00:01:02.009
<v Speaker 1>and lowered as plots into the ocean to grow, which

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:02.020 --> 00:01:04.290
<v Speaker 1>can then be pulled up to be harvested later on

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:04.989 --> 00:01:06.849
<v Speaker 1>to answer this question of habitat.

NOTE CONF {"raw":[99,99,99,99,99,99]}

00:01:07.080 --> 00:01:10.910
<v Speaker 1>We put down underwater video cameras in seaweed farms here

NOTE CONF {"raw":[99,99,99,98,99,99,99,99,99,99]}

00:01:10.919 --> 00:01:14.919
<v Speaker 1>in BC and recorded over 9000 hours of video over

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:14.930 --> 00:01:18.419
<v Speaker 1>16 months to find out, do we see more fish

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:18.430 --> 00:01:21.330
<v Speaker 1>and other sea life in seaweed farms than we do

NOTE CONF {"raw":[99,99,99,96,99,99,99,88,99,99]}

00:01:21.339 --> 00:01:23.730
<v Speaker 1>in surrounding areas without seaweed.

NOTE CONF {"raw":[99,99,99,99,99]}

00:01:24.150 --> 00:01:28.000
<v Speaker 1>But as you may expect, looking through 9000 hours of

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:28.010 --> 00:01:29.930
<v Speaker 1>video would be extremely challenging.

NOTE CONF {"raw":[99,99,99,99,99]}

00:01:30.139 --> 00:01:32.330
<v Speaker 1>So this is where my thesis comes in.

NOTE CONF {"raw":[83,99,99,99,99,99,99,99]}

00:01:32.819 --> 00:01:36.319
<v Speaker 1>My work is to reduce the effort involved in processing

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:36.330 --> 00:01:40.940
<v Speaker 1>these videos by training artificial intelligence to essentially become a

NOTE CONF {"raw":[99,99,86,99,99,99,99,99,99,99]}

00:01:40.949 --> 00:01:42.400
<v Speaker 1>robot biologist.

NOTE CONF {"raw":[99,99]}

00:01:42.839 --> 00:01:45.410
<v Speaker 1>What we do is take a small portion of all

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:45.419 --> 00:01:46.459
<v Speaker 1>videos collected.

NOTE CONF {"raw":[99,99]}

00:01:46.650 --> 00:01:49.769
<v Speaker 1>We identify every animal in each of those videos and

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:49.779 --> 00:01:52.639
<v Speaker 1>then we use that information to train a custom built

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,68]}

00:01:52.650 --> 00:01:56.919
<v Speaker 1>computer program to also become really good at identifying these

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:01:56.930 --> 00:01:57.919
<v Speaker 1>same species.

NOTE CONF {"raw":[99,99]}

00:01:58.430 --> 00:02:01.059
<v Speaker 1>And once we're confident in its ability, we can set

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:01.069 --> 00:02:03.919
<v Speaker 1>it loose on the entire video data set to answer

NOTE CONF {"raw":[99,99,99,99,99,99,99,96,99,99]}

00:02:03.930 --> 00:02:04.489
<v Speaker 1>our question.

NOTE CONF {"raw":[99,99]}

00:02:05.110 --> 00:02:08.169
<v Speaker 1>And using this method, we've already had pretty good results.

NOTE CONF {"raw":[99,99,99,99,78,99,99,99,99,99]}

00:02:08.570 --> 00:02:11.139
<v Speaker 1>This fish school you see up on the screen was

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:11.149 --> 00:02:14.589
<v Speaker 1>correctly plucked from raw video data and identified to a

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:14.600 --> 00:02:18.089
<v Speaker 1>species pacific herring entirely by a computer.

NOTE CONF {"raw":[99,99,99,85,99,99,99]}

00:02:18.910 --> 00:02:21.850
<v Speaker 1>And we've trained the same same program to identify over

NOTE CONF {"raw":[99,99,99,99,87,99,99,99,99,99]}

00:02:21.860 --> 00:02:25.210
<v Speaker 1>20 different types of fish, jellyfish and other marine life

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:25.220 --> 00:02:27.529
<v Speaker 1>found right here in the s of sea.

NOTE CONF {"raw":[99,99,99,99,99,56,0,99]}

00:02:28.279 --> 00:02:31.440
<v Speaker 1>This is really exciting because not only can this help

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:31.449 --> 00:02:34.320
<v Speaker 1>us answer our question of seaweed farm impacts on marine

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:34.330 --> 00:02:37.179
<v Speaker 1>life, but it can create a tool that can support

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:37.190 --> 00:02:43.070
<v Speaker 1>a future with more accessible, robust and accurate underwater biodiversity

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,98,99]}

00:02:43.080 --> 00:02:46.009
<v Speaker 1>data, which can help us tackle big questions in our

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:46.020 --> 00:02:50.440
<v Speaker 1>rapidly changing ocean and it can empower industry to better

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:50.449 --> 00:02:55.160
<v Speaker 1>monitor the impacts of their own operations, impacts like and

NOTE CONF {"raw":[99,99,99,99,99,99,99,99,99,99]}

00:02:55.169 --> 00:02:58.199
<v Speaker 1>questions like the one we're tackling with this project.

NOTE CONF {"raw":[99,99,99,99,98,99,99,99,99]}

00:02:58.399 --> 00:03:01.119
<v Speaker 1>What is going on on the farm?

NOTE CONF {"raw":[99,99,99,99,99,99,99]}

00:03:01.479 --> 00:03:01.919
<v Speaker 1>Thank you.

NOTE CONF {"raw":[99,99]}