Skip to content

Commit 15dceb6

Browse files
anish-lakkapragadaanish-lakkapragada
committed
web
1 parent 5b53f45 commit 15dceb6

16 files changed

+3192
-107
lines changed

‎Approach.png‎

44.7 KB
Loading

‎README.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -127,6 +127,7 @@ The <code>ensemble_code</code> folder contains the notebooks in which we tried t
127127
## Citations
128128
Please cite the following:
129129
```
130+
130131
@article{lakkapragada2021activity,
131132
title={Activity Recognition for Autism Diagnosis},
132133
author={Lakkapragada, Anish and Washington, Peter and Wall, Dennis},

‎index.html‎

Lines changed: 108 additions & 107 deletions
Original file line numberDiff line numberDiff line change
@@ -3,9 +3,7 @@
33

44
<head>
55
<meta charset="utf-8">
6-
<meta name="description"
7-
content="Activity Recognition for Autism Diagnosis">
8-
<meta name="keywords" content="Nerfies, D-NeRF, NeRF">
6+
<meta name="description" content="Activity Recognition for Autism Diagnosis">
97
<meta name="viewport" content="width=device-width, initial-scale=1">
108
<title>Activity Recognition for Autism Diagnosis</title>
119

@@ -43,28 +41,6 @@
4341
</head>
4442

4543
<body>
46-
47-
<nav class="navbar" role="navigation" aria-label="main navigation">
48-
<div class="navbar-brand">
49-
<a role="button" class="navbar-burger" aria-label="menu" aria-expanded="false">
50-
<span aria-hidden="true"></span>
51-
<span aria-hidden="true"></span>
52-
<span aria-hidden="true"></span>
53-
</a>
54-
</div>
55-
<div class="navbar-menu">
56-
<div class="navbar-start" style="flex-grow: 1; justify-content: center;">
57-
<a class="navbar-item" href="https://keunhong.com">
58-
<span class="icon">
59-
<i class="fas fa-home"></i>
60-
</span>
61-
</a>
62-
</div>
63-
64-
</div>
65-
</nav>
66-
67-
6844
<section class="hero">
6945
<div class="hero-body">
7046
<div class="container is-max-desktop">
@@ -73,17 +49,20 @@
7349
<h1 class="title is-1 publication-title"> Activity Recognition for Autism Diagnosis</h1>
7450
<div class="is-size-5 publication-authors">
7551
<span class="author-block">
76-
<a href="https://keunhong.com">Anish Lakkapragada</a><sup>1</sup>,</span>
52+
<a href="https://wall-lab.stanford.edu/people/interns/anishl/">Anish
53+
Lakkapragada</a><sup>1</sup>,</span>
7754
<span class="author-block">
78-
<a href="https://utkarshsinha.com">Peter Washington</a><sup>2</sup>,</span>
55+
<a href="https://wall-lab.stanford.edu/people/current/peter/">Peter
56+
Washington</a><sup>2</sup>,</span>
7957
<span class="author-block">
80-
<a href="https://jonbarron.info">Dennis P. Wall</a><sup>2</sup>,
58+
<a href="https://wall-lab.stanford.edu/people/current/dennis/">Dennis P.
59+
Wall</a><sup>2</sup>,
8160
</span>
8261
</div>
8362

8463
<div class="is-size-5 publication-authors">
85-
<span class="author-block"><sup>1</sup>Lynbrook High School</span>
86-
<span class="author-block"><sup>2</sup>Stanford University</span>
64+
<span class="author-block">Lynbrook High School<sup>1</sup></span>
65+
<span class="author-block">Stanford University<sup>2</sup></span>
8766
</div>
8867

8968
<div class="column has-text-centered">
@@ -97,7 +76,7 @@ <h1 class="title is-1 publication-title"> Activity Recognition for Autism Diagno
9776
</span>
9877
<span>Paper</span>
9978
</a>
100-
</span> -->
79+
</span> -->
10180
<span class="link-block">
10281
<a href="https://arxiv.org/abs/2108.07917"
10382
class="external-link button is-normal is-rounded is-dark">
@@ -116,7 +95,7 @@ <h1 class="title is-1 publication-title"> Activity Recognition for Autism Diagno
11695
</span>
11796
<span>Video</span>
11897
</a>
119-
</span> -->
98+
</span> -->
12099
<!-- Code Link. -->
121100

122101
<span class="link-block">
@@ -151,7 +130,7 @@ <h2 class="subtitle has-text-centered">
151130
</h2>
152131
</div>
153132
</div>
154-
</section>-->
133+
</section>-->
155134

156135
<!--
157136
<section class="hero is-light is-small">
@@ -210,99 +189,121 @@ <h2 class="subtitle has-text-centered">
210189
</div>
211190
</div>
212191
</section>
213-
-->
192+
-->
214193

215194
<section class="section">
216195
<div class="container is-max-desktop">
217-
<!-- Abstract. -->
196+
197+
<!-- Overview-->
198+
218199
<div class="columns is-centered has-text-centered">
219200
<div class="column is-four-fifths">
220-
<h2 class="title is-3">Abstract</h2>
221-
<div class="content has-text-justified">
222-
<p> <em> A formal autism diagnosis can be an inefficient and lengthy process. Families may wait months orlonger before
223-
receiving a diagnosis for their child. One approach to lessen delays is the use digitaltechnologies to detect
224-
the presence of behaviors indicative of autism, which in aggregate may leadto remote and automated diagnostics.
225-
One of the strongest indicators of autism is stimming, whichincludes repetitive, self-stimulatory behaviors such
226-
as hand flapping, headbanging, and spinning.Using computer vision to detect hand flapping is especially
227-
difficult due to the sparsity of publictraining data in this space and excessive shakiness and motion in such
228-
data. Our work demonstratesa novel method that may overcome these issues: we use hand landmark detection over
229-
time as afeature representation which is then fed into a Long Short-Term Memory (LSTM) model. We achieve
230-
avalidation accuracy and F1 Score of about 72% on detecting whether videos from the Self-StimulatoryBehaviour
231-
Dataset (SSBD) contain hand flapping or not. Our best model also predicts accurately onexternal videos we
232-
recorded of ourselves outside of the dataset it was trained on. This model usesless than 26,000 parameters,
233-
providing promise for fast deployment into ubiquitous and wearabledigital settings for a remote autism
234-
diagnosis. </em> </p>
201+
<h2 class="title is-3">Overview</h2>
202+
<div class="content has-text-justified">
203+
<figure>
204+
<img src = "Approach.png">
205+
<figcaption> Our overall model to detect hand flapping, an indicator of autism, in videos. </figcaption>
206+
</figure>
207+
<p> Our objective was to be able to create a model that could reliably detect hand flapping, an indicator of autism, in videos that then could be applied
208+
to apps to help create a remote autism diagnosis. We do this by going frame (an image) by frame in a given video, and detecting the
209+
numerical coordinates of the hand's landmarks in the images. These coordinates are fed overtime into a Long Term Short-Term Memory (LSTM) model. The output
210+
at the last time step of the LSTM model is fed into a fully-connected layer to get the prediction on whether hand flapping was detected (1) or not (0). A visual diagram
211+
of our approach is shown above. Our best model uses less than 26,000 parameters, so it can easily be deployed into any app.</p>
235212
</div>
236213
</div>
237214
</div>
238-
<!--/ Abstract. -->
239215

240-
<!-- Paper video.
216+
<!-- Abstract. -->
241217
<div class="columns is-centered has-text-centered">
242218
<div class="column is-four-fifths">
243-
<h2 class="title is-3">Video</h2>
244-
<div class="publication-video">
245-
<iframe src="https://www.youtube.com/embed/MrKrnHhk8IA?rel=0&amp;showinfo=0" frameborder="0"
246-
allow="autoplay; encrypted-media" allowfullscreen></iframe>
219+
<h2 class="title is-3">Abstract</h2>
220+
<div class="content has-text-justified">
221+
<p> <em> A formal autism diagnosis can be an inefficient and lengthy process. Families may wait
222+
months orlonger before
223+
receiving a diagnosis for their child. One approach to lessen delays is the use
224+
digitaltechnologies to detect
225+
the presence of behaviors indicative of autism, which in aggregate may leadto remote and
226+
automated diagnostics.
227+
One of the strongest indicators of autism is stimming, whichincludes repetitive,
228+
self-stimulatory behaviors such
229+
as hand flapping, headbanging, and spinning.Using computer vision to detect hand
230+
flapping is especially
231+
difficult due to the sparsity of publictraining data in this space and excessive
232+
shakiness and motion in such
233+
data. Our work demonstratesa novel method that may overcome these issues: we use hand
234+
landmark detection over
235+
time as afeature representation which is then fed into a Long Short-Term Memory (LSTM)
236+
model. We achieve
237+
avalidation accuracy and F1 Score of about 72% on detecting whether videos from the
238+
Self-StimulatoryBehaviour
239+
Dataset (SSBD) contain hand flapping or not. Our best model also predicts accurately
240+
onexternal videos we
241+
recorded of ourselves outside of the dataset it was trained on. This model usesless than
242+
26,000 parameters,
243+
providing promise for fast deployment into ubiquitous and wearabledigital settings for a
244+
remote autism
245+
diagnosis. </em> </p>
247246
</div>
248247
</div>
249248
</div>
250-
-->
251249

252250

253-
</div>
254-
</section>
255-
256-
257-
<section class="section">
258-
<div class="container is-max-desktop"></div>
259-
</section>
260-
261-
262-
<section class="section" id="BibTeX">
263-
<div class="container is-max-desktop content">
264-
<h2 class="title">BibTeX</h2>
265-
<pre><code>@article{park2021nerfies
266-
author = {Park, Keunhong and Sinha, Utkarsh and Barron, Jonathan T. and Bouaziz, Sofien and Goldman, Dan B and Seitz, Steven M. and Martin-Brualla, Ricardo},
267-
title = {Nerfies: Deformable Neural Radiance Fields},
268-
journal = {ICCV},
269-
year = {2021},
270-
}</code></pre>
271-
</div>
272-
</section>
273-
274-
275-
<footer class="footer">
276-
<div class="container">
277-
<div class="content has-text-centered">
278-
<a class="icon-link" href="https://homes.cs.washington.edu/~kpar/nerfies/videos/nerfies_paper.pdf">
279-
<i class="fas fa-file-pdf"></i>
280-
</a>
281-
<a class="icon-link" href="https://github.com/keunhong" class="external-link" disabled>
282-
<i class="fab fa-github"></i>
283-
</a>
284-
</div>
285-
<div class="columns is-centered">
286-
<div class="column is-8">
287-
<div class="content">
288-
<p>
289-
This website is licensed under a <a rel="license"
290-
href="http://creativecommons.org/licenses/by-sa/4.0/">Creative
291-
Commons Attribution-ShareAlike 4.0 International License</a>.
292-
</p>
293-
<p>
294-
This means you are free to borrow the <a
295-
href="https://github.com/nerfies/nerfies.github.io">source code</a> of this website,
296-
we just ask that you link back to this page in the footer.
297-
Please remember to remove the analytics code included in the header of the website which
298-
you do not want on your website.
299-
</p>
251+
252+
<section class="section">
253+
<div class="container-fluid">
254+
<div class="columns is-centered has-text-centered">
255+
<div class="column is-two-thirds">
256+
<h2 class="title is-2">Paper</h2>
257+
<div class="content has-text-justified">
258+
259+
<embed src="https://arxiv.org/pdf/2108.07917.pdf" width="100%" height='700px' />
260+
261+
</div>
262+
</div>
300263
</div>
301264
</div>
302-
</div>
265+
</section>
266+
267+
268+
<section class="section" id="BibTeX">
269+
<div class="container is-max-desktop content">
270+
<h2 class="title">Citation</h2>
271+
<pre> <code>
272+
@article{lakkapragada2021activity,
273+
title={Activity Recognition for Autism Diagnosis},
274+
author={Lakkapragada, Anish and Washington, Peter and Wall, Dennis},
275+
journal={arXiv preprint arXiv:2108.07917},
276+
year={2021}
277+
}
278+
</code> </pre>
279+
</div>
280+
</section>
281+
282+
283+
<footer class="footer">
284+
<div class="container">
285+
<div class="content has-text-centered">
286+
<a class="icon-link" href="static/paper.pdf">
287+
<i class="fas fa-file-pdf"></i>
288+
</a>
289+
<a class="icon-link" href="https://github.com/ayaanzhaque" class="external-link" disabled>
290+
<i class="fab fa-github"></i>
291+
</a>
292+
</div>
293+
<div class="columns is-centered">
294+
<div class="column is-8">
295+
<div class="content">
296+
<p>
297+
This website template was used from <a
298+
href="https://github.com/nerfies/nerfies.github.io">here</a>.
299+
</p>
300+
</div>
301+
</div>
302+
</div>
303+
</div>
304+
</footer>
303305
</div>
304-
</footer>
305-
306+
</section>
306307
</body>
307308

308309
</html>

‎static/css/bulma-carousel.min.css‎

Lines changed: 1 addition & 0 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

0 commit comments

Comments
 (0)