|
|
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3c.org/TR/1999/REC-html401-19991224/loose.dtd"> |
|
|
<html xml:lang="en" xmlns="http://www.w3.org/1999/xhtml" lang="en"><head> |
|
|
<title>The iNaturalist Sounds Dataset</title> |
|
|
<meta http-equiv="Content-Type" content="text/html; charset=windows-1252"> |
|
|
|
|
|
<meta property="og:title" content="The iNaturalist Sounds Dataset" /> |
|
|
|
|
|
<script> |
|
|
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ |
|
|
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), |
|
|
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) |
|
|
})(window,document,'script','https://www.google-analytics.com/analytics.js','ga'); |
|
|
|
|
|
ga('create', 'UA-167124286-2', 'auto'); |
|
|
ga('send', 'pageview'); |
|
|
</script> |
|
|
|
|
|
|
|
|
|
|
|
<script type="text/javascript"> |
|
|
|
|
|
var _POPUP_FEATURES = 'width=500,height=300,resizable=1,scrollbars=1,titlebar=1,status=1'; |
|
|
</script> |
|
|
|
|
|
<style type="text/css" media="all"> |
|
|
IMG { |
|
|
PADDING-RIGHT: 0px; |
|
|
PADDING-LEFT: 0px; |
|
|
FLOAT: right; |
|
|
PADDING-BOTTOM: 0px; |
|
|
PADDING-TOP: 0px |
|
|
} |
|
|
#primarycontent { |
|
|
MARGIN-LEFT: auto; ; WIDTH: expression(document.body.clientWidth > |
|
|
1000? "1000px": "auto" ); MARGIN-RIGHT: auto; TEXT-ALIGN: left; max-width: |
|
|
1000px } |
|
|
BODY { |
|
|
TEXT-ALIGN: center |
|
|
} |
|
|
</style> |
|
|
|
|
|
<style type="text/css"> |
|
|
body { |
|
|
font-family: "Titillium Web","HelveticaNeue-Light", "Helvetica Neue Light", "Helvetica Neue", Helvetica, Arial, "Lucida Grande", sans-serif; |
|
|
font-weight:300; |
|
|
font-size:18px; |
|
|
margin-left: auto; |
|
|
margin-right: auto; |
|
|
margin-bottom: 0px; |
|
|
width: 100%; |
|
|
} |
|
|
|
|
|
h1 { |
|
|
font-weight:300; |
|
|
} |
|
|
|
|
|
div { |
|
|
max-width: 95%; |
|
|
margin:auto; |
|
|
padding: 10px; |
|
|
} |
|
|
|
|
|
.table-like { |
|
|
display: flex; |
|
|
flex-wrap: wrap; |
|
|
flex-flow: row wrap; |
|
|
justify-content: center; |
|
|
} |
|
|
|
|
|
.disclaimerbox { |
|
|
background-color: #eee; |
|
|
border: 1px solid #eeeeee; |
|
|
border-radius: 10px ; |
|
|
-moz-border-radius: 10px ; |
|
|
-webkit-border-radius: 10px ; |
|
|
padding: 20px; |
|
|
} |
|
|
|
|
|
video.header-vid { |
|
|
height: 140px; |
|
|
border: 1px solid black; |
|
|
border-radius: 10px ; |
|
|
-moz-border-radius: 10px ; |
|
|
-webkit-border-radius: 10px ; |
|
|
} |
|
|
|
|
|
img { |
|
|
padding: 0; |
|
|
display: block; |
|
|
margin: 0 auto; |
|
|
max-height: 100%; |
|
|
max-width: 100%; |
|
|
} |
|
|
|
|
|
iframe { |
|
|
max-width: 100%; |
|
|
} |
|
|
|
|
|
img.header-img { |
|
|
height: 140px; |
|
|
border: 1px solid black; |
|
|
border-radius: 10px ; |
|
|
-moz-border-radius: 10px ; |
|
|
-webkit-border-radius: 10px ; |
|
|
} |
|
|
|
|
|
img.rounded { |
|
|
border: 1px solid #eeeeee; |
|
|
border-radius: 10px ; |
|
|
-moz-border-radius: 10px ; |
|
|
-webkit-border-radius: 10px ; |
|
|
} |
|
|
|
|
|
a:link,a:visited |
|
|
{ |
|
|
color: #1367a7; |
|
|
text-decoration: none; |
|
|
} |
|
|
a:hover { |
|
|
color: #208799; |
|
|
} |
|
|
|
|
|
td.dl-link { |
|
|
height: 160px; |
|
|
text-align: center; |
|
|
font-size: 22px; |
|
|
} |
|
|
|
|
|
.layered-paper-big { |
|
|
box-shadow: |
|
|
0px 0px 1px 1px rgba(0,0,0,0.35), |
|
|
5px 5px 0 0px #fff, |
|
|
5px 5px 1px 1px rgba(0,0,0,0.35), |
|
|
10px 10px 0 0px #fff, |
|
|
10px 10px 1px 1px rgba(0,0,0,0.35), |
|
|
15px 15px 0 0px #fff, |
|
|
15px 15px 1px 1px rgba(0,0,0,0.35), |
|
|
20px 20px 0 0px #fff, |
|
|
20px 20px 1px 1px rgba(0,0,0,0.35), |
|
|
25px 25px 0 0px #fff, |
|
|
25px 25px 1px 1px rgba(0,0,0,0.35); |
|
|
margin-left: 10px; |
|
|
margin-right: 45px; |
|
|
} |
|
|
|
|
|
|
|
|
.layered-paper { |
|
|
box-shadow: |
|
|
0px 0px 1px 1px rgba(0,0,0,0.35), |
|
|
5px 5px 0 0px #fff, |
|
|
5px 5px 1px 1px rgba(0,0,0,0.35), |
|
|
10px 10px 0 0px #fff, |
|
|
10px 10px 1px 1px rgba(0,0,0,0.35); |
|
|
margin-top: 5px; |
|
|
margin-left: 10px; |
|
|
margin-right: 30px; |
|
|
margin-bottom: 5px; |
|
|
} |
|
|
|
|
|
.vert-cent { |
|
|
position: relative; |
|
|
top: 50%; |
|
|
transform: translateY(-50%); |
|
|
} |
|
|
|
|
|
hr |
|
|
{ |
|
|
border: 0; |
|
|
height: 1px; |
|
|
max-width: 1100px; |
|
|
background-image: linear-gradient(to right, rgba(0, 0, 0, 0), rgba(0, 0, 0, 0.75), rgba(0, 0, 0, 0)); |
|
|
} |
|
|
|
|
|
#authors td { |
|
|
padding-bottom:5px; |
|
|
padding-top:30px; |
|
|
} |
|
|
|
|
|
.code-block{ |
|
|
font-size:14px; |
|
|
text-align: left; |
|
|
max-width:700px; |
|
|
overflow-x: scroll; |
|
|
background-color:azure; |
|
|
border-color: rgb(192, 192, 192); |
|
|
border-width: 2px; |
|
|
border-radius: 10px; |
|
|
border-style:solid; |
|
|
} |
|
|
.text-block{ |
|
|
font-size:16px; |
|
|
text-align: justify; |
|
|
max-width:750px |
|
|
} |
|
|
</style> |
|
|
|
|
|
<script> |
|
|
function resizeIframe(obj) { |
|
|
obj.style.height = obj.contentWindow.document.documentElement.scrollHeight + 'px'; |
|
|
} |
|
|
</script> |
|
|
|
|
|
<body> |
|
|
|
|
|
<div id="primarycontent"> |
|
|
<center><h1 style="font-size: 225%">The iNaturalist Sounds Dataset</h1></center> |
|
|
<center> |
|
|
|
|
|
<div class="table-like" style="max-width:800px;margin:auto;"> |
|
|
|
|
|
<div> |
|
|
<center> |
|
|
<a href="https://mustafa1728.github.io" target="_blank" style="font-size: larger">Mustafa Chasmai</a> |
|
|
</center> |
|
|
<center> |
|
|
UMass Amherst |
|
|
</center> |
|
|
</div> |
|
|
|
|
|
<div> |
|
|
<center> |
|
|
<a href="https://www.inaturalist.org/users/44845" target="_blank" style="font-size: larger">Alexander Shepard</a> |
|
|
</center> |
|
|
<center> |
|
|
iNaturalist |
|
|
</center> |
|
|
</div> |
|
|
|
|
|
<div> |
|
|
<center> |
|
|
<a href="https://people.cs.umass.edu/~smaji/" target="_blank" style="font-size: larger">Subhransu Maji</a> |
|
|
</center> |
|
|
<center> |
|
|
UMass Amherst |
|
|
</center> |
|
|
</div> |
|
|
|
|
|
<div> |
|
|
<center> |
|
|
<a href="https://gvh.codes/" target="_blank" style="font-size: larger">Grant Van Horn</a> |
|
|
</center> |
|
|
<center> |
|
|
UMass Amherst |
|
|
</center> |
|
|
</div> |
|
|
|
|
|
</div> |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<div class="table-like" style="justify-content:space-evenly;max-width:900px;margin:auto;"> |
|
|
<center> |
|
|
<table> |
|
|
<tr> |
|
|
<td style="font-size:20px;margin:20px;font-family:monospace"> |
|
|
<a style="margin:2px" href="https://openreview.net/pdf?id=QCY01LvyKm">[Paper]</a> |
|
|
</td> |
|
|
<td style="font-size:20px;margin:20px;font-family:monospace"> |
|
|
<a style="margin:2px" href="https://github.com/cvl-umass/iNatSounds">[Code]</a> |
|
|
</td> |
|
|
<td style="font-size:20px;margin:20px;font-family:monospace"> |
|
|
<a style="margin:2px" href="https://github.com/visipedia/inat_sounds">[Data]</a> |
|
|
</td> |
|
|
</tr> |
|
|
</table> |
|
|
</center> |
|
|
</div> |
|
|
<center> |
|
|
<br> |
|
|
|
|
|
|
|
|
<center> |
|
|
<table border="0" cellspacing="0" cellpadding="0"> |
|
|
<tr> |
|
|
<td align="center" valign="bottom" style="max-width:800px"> |
|
|
<img class="result" src="assets/overview.jpg" style="width: 100%"> |
|
|
</td> |
|
|
</tr> |
|
|
</table> |
|
|
</center> |
|
|
|
|
|
<h2>Abstract</h2> |
|
|
<div style="font-size:16px; text-align: justify;max-width:700px"> |
|
|
<p> |
|
|
We present the iNaturalist Sounds Dataset (iNatSounds), a collection of 230,000 audio files capturing sounds from over 5,500 species, contributed by more than 27,000 recordists worldwide. |
|
|
The dataset encompasses sounds from birds, mammals, insects, reptiles, and amphibians, with audio and species labels derived from observations submitted to iNaturalist, a global citizen science platform. |
|
|
Each recording in the dataset varies in length and includes a single species annotation. |
|
|
We benchmark multiple backbone architectures, comparing multiclass classification objectives with multilabel objectives. |
|
|
Despite weak labeling, we demonstrate that iNatSounds serves as a robust pretraining resource, achieving high performance relative to alternatives on strongly labeled downstream evaluation datasets. |
|
|
The dataset is available as a single, freely accessible archive, promoting accessibility and research in this important domain. |
|
|
We envision models trained on this data powering next-generation public engagement applications, and assisting biologists, ecologists, and land use managers in processing large audio collections, thereby contributing to the understanding of species compositions in diverse soundscapes. |
|
|
</p> |
|
|
</div> |
|
|
|
|
|
|
|
|
<center> |
|
|
<table border="0" cellspacing="0" cellpadding="0"> |
|
|
<tr> |
|
|
<td align="center" valign="bottom" style="max-width:800px"> |
|
|
<img class="result" src="assets/class_counts.jpg" style="width:100%"> |
|
|
</td> |
|
|
</tr> |
|
|
</table> |
|
|
</center> |
|
|
|
|
|
<h2>Citation</h2> |
|
|
<pre style="font-size:14px; text-align: left;max-width:700px;overflow-x: scroll;background-color:beige;border-color: rgb(192, 192, 192);border-width: 2px;border-radius: 10px;border-style:solid;"> |
|
|
<code> |
|
|
@article{chasmai2024inaturalist, |
|
|
title={The iNaturalist Sounds Dataset}, |
|
|
author={Chasmai, Mustafa and Shepard, Alex and Maji, Subhransu and Van Horn, Grant}, |
|
|
journal={Advances in Neural Information Processing Systems}, |
|
|
year={2024} |
|
|
} |
|
|
</code> |
|
|
</pre> |
|
|
|
|
|
|
|
|
</body> |
|
|
|
|
|
</html> |
|
|
|
