Computer Vision for the Environment,
Sustainability, and Conservation
I'm an assistant professor at MIT EECS' Faculty of AI and Decision Making and CSAIL, and I was previously a visiting researcher at Google working on Auto Arborist. I’ve always loved the natural world, and I've seen a growing need for technology-based approaches to conservation and sustainability challenges. My research focuses on building computer vision methods that enable global-scale environmental and biodiversity monitoring across data modalities, tackling real-world challenges including strong spatiotemporal correlations that lead to domain shift, imperfect data quality, fine-grained categories, and long-tailed distributions. I received my PhD in Computing and Mathematical Sciences (CMS) at Caltech, advised by Pietro Perona, where I received the Amori Doctoral Prize for my dissertation. I have been honored to be awarded an AI2050 Early Career Fellowship, a PIMCO Data Science Fellowship an Amazon AI4Science Fellowship, and the NSF Graduate Research Fellowship. My work has been supported by the NSF, NASA, Google, Microsoft, IBM, the USAF, MIT J-WAFS, and the Caltech Resnick Sustainability Institute.
I seek to break down knowledge barriers between fields: I founded the successful AI for Conservation slack community (with over 2500 members), I am Co-PI on the NSF/NSERC Global Center on AI and Biodiversity Change, I am the Biodiversity Community Lead for Climate Change AI, and I am the founding director of the Caltech Summer School on Computer Vision Methods for Ecology. I work closely with industry, including the MIT-IBM Watson AI Lab, Microsoft AI for Good, Google Research, Wildlife Insights, and Wild Me where I help turn my research into usable, deployed tools.
My experiences as a professional ballerina and a nontraditional student have taught me the value of unique perspectives, both inside and outside of the research community. I am passionate about increasing access to and capacity in STEM through mentorship, teaching, and outreach, and was honored to have my efforts recognized with the inaugural Caltech Computing and Mathematical Sciences Department Gradient for Change Award and the inaugural Caltech Engineering and Applied Science Division New Horizons Award.
PhD Students
Julia Chae
Timm Haucke
Neha Hulkund
Justin Kay
Eddie Vendrow
Rupa Kurinchi-Vendhan
Masters and Visiting Students
Lauren Shrack
Antoine Salaun
Undergraduates
Kai Van Brunt
Xenia Zhao
Asmi Kumar
David Fang (Alumni)
Ari Pero (Alumni)
Avi Sundaresan (Caltech)
If you are interested in any position with the group (Postdoc, PhD, MEng, UROP, Visitor), instead of emailing me please fill out the interest survey below so I can better organize my responses. I am unlikely to respond to individual emails.
Interested in a PhD? I consider students primarily via MIT EECS and the MIT-WHOI joint program, but I would be open to advising students with the right fit for my lab through MIT EAPS or MIT CEE if that is better aligned with your background. If you are interested in working with me on impact-driven computer vision research for the environment, biodiversity, conservation, and sustainability, apply to any of these programs and list me in your application. Unfortunately I am unlikely to be able to meet with prospective applicants before they have gone through the formal review process.
Interested in a postdoc? I will be welcoming postdocs on an ongoing basis, depending on funding availability. If our research interests align, please fill out the interest survey below and include any ideas you have for projects you would be interested in exploring together. I am happy to develop fellowship proposal(s) together. I'm particularly interested in bringing on a postdoctoral researcher with a background in quantitative large-scale biodiversity monitoring to explore the intersection between machine learning outputs and ecological modeling inputs with me and my group.
This list is a selected subset and may not be up to date. For a full list of publications please see my CV or Google Scholar. * denotes equal contribution, ^ denotes equal advising
Application-Driven Innovation in Machine Learning
David Rolnick, Alan Aspuru-Guzik, Sara Beery, Bistra Dilkina, Priya L. Donti, Marzyeh Ghassemi, Hannah Kerner, Claire Monteleoni, Esther Rolf, Milind Tambe, Adam White
[paper]
Align and Distill: Unifying and Improving Domain Adaptive Object Detection
Justin Kay, Timm Haucke, Suzanne Stathatos, Siqi Deng, Erik Young, Pietro Perona, Sara Beery^, Grant Van Horn^
[paper]
[codebase]
[dataset]
Sustainable, trustworthy, human-technology partnership
Tanya Berger-Wolf, Sara Beery, David Rolnick, Justin Kitzes, David Thau, Devis Tuia, Daniel Rubenstein
A landmark environmental law looks ahead.
Science, 2023
[paper]
Teaching Computer Vision for Ecology.
Elijah Cole, Suzanne Stathatos, Björn Lütjens, Tarun Sharma, Justin Kay, Jason Parham, Benjamin Kellenberger, and Sara Beery.
[paper]
[cv4ecology workshop]
.png)
Perspectives in Machine Learning for Wildlife Conservation
Devis Tuia*, Benjamin Kellenberger*, Sara Beery*, Blair R. Costelloe*, Silvia Zuffi, Benjamin Risse, Alexander Mathis, Mackenzie W. Mathis, Frank van Langevelde, Tilo Burghardt, Roland Kays, Holger Klinck, Martin Wikelski, Iain D. Couzin, Grant van Horn, Margaret C. Crofoot, Charles V. Stewart, Tanya Berger-Wolf
Nature Communications 2022
[paper]
Extending the WILDS Benchmark for Unsupervised Adaptation
Shiori Sagawa, Pang Wei Koh, Tony Lee, Irena Gao, Sang Michael Xie, Kendrick Shen, Ananya Kumar, Weihua Hu, Michihiro Yasunaga, Henrik Marklund, Sara Beery, Etienne David, Ian Stavness, Wei Guo, Jure Leskovec, Kate Saenko, Tatsunori Hashimoto, Sergey Levine, Chelsea Finn, Percy Liang
ICLR 2022 (Oral)
[paper][code/data]
Scaling Biodiversity Monitoring for the Data Age
Sara Beery
ACM XRDS 2021
[paper]
Wilds: A benchmark of in-the-wild distribution shifts
Pang Wei Koh, Shiori Sagawa, Sang Michael Xie, Marvin Zhang, Akshay Balsubramani, Weihua Hu, Michihiro Yasunaga, Richard Lanas Phillips, Irena Gao, Tony Lee, Etienne David, Ian Stavness, Wei Guo, Berton Earnshaw, Imran Haque, Sara Beery, Jure Leskovec, Anshul Kundaje, Emma Pierson, Sergey Levine, Chelsea Finn, Percy Liang
ICLR 2021 (Oral)
[paper][code/data]
Benchmarking Representation Learning for Natural World Image Collections
Grant Van Horn, Elijah Cole, Sara Beery, Kimberly Wilber, Serge Belongie, and Oisin Mac Aodha
CVPR 2021 (Oral)
[paper][code/data][video]
Species Distribution Modeling for Machine Learning Practitioners: A Review
Sara Beery*, Elijah Cole*, Joseph Parker, Pietro Perona, Kevin Winner
ACM COMPASS 2021
[paper]
ElephantBook: A Semi-Automated Human-in-the-Loop System for Elephant Re-Identification
Peter Kulits, Jake Wall, Anka Bedetti, Michelle Henley, Sara Beery
ACM COMPASS 2021
[paper]
A Deep Active Learning System for Species Identification and Counting in Camera Trap Images
Mohammad Sadegh Norouzzadeh, Dan Morris, Sara Beery, Neel Joshi, Nebojsa Jojic, Jeff Clune
Methods in Ecology and Evolution 2021
[paper]
Synthetic Examples Improve Generalization for Rare Classes
Sara Beery, Yang Liu, Dan Morris, Jim Piavis, Ashish Kapoor, Markus Meister, Neel Joshi, Pietro Perona
WACV 2020
[paper]
Finding Areas of Motion in Camera Trap Images
Agnieszka Miguel, Sara Beery, Erica Flores, Loren Klemesrud
IEEE ICIP 2016
[paper]
In Summer 2022 I will be directing the first annual Resnick Sustainability Institute Summer School on Computer Vision Methods for Ecology (https://cv4ecology.caltech.edu/). This intensive, three-week program will teach applied computer vision methods to senior ecology graduate students and postdocs.
Students will develop hands-on computer vision systems to help answer their own ecological research questions, using their own data. They will receive daily mentorship from a passionate team of computer vision experts with a track record of impact in conservation and sustainability. Each student will be provided with $2500 in cloud credits to facilitate their project development sponsored by Microsoft AI for Earth and Amazon AWS.
Our team of instructors will work with applicants leading up to the intensive to curate computer-vision-ready labels for their data that will be used to prototype systems for their research questions during the class. Students will leave the course empowered to build their own computer vision models for ecological applications, and gain skills in problem formulation, dataset curation, model training, model evaluation, and hosting models for inference.
I helped adapt this advanced projects-based computer vision course to focus on conservation and sustainability applications. To set up project groups for success, I curated a set of ecological challenges with publicly available image and video datasets and matched projects to NGOs and research groups that would directly benefit to provide domain expertise and context. I mentored 5 teams of computer vision students in structuring and defining these real-world challenges as computer vision and machine learning problems, and assisted them in holistically probing and evaluating their solutions and effectively communicating them to both computer vision and machine learning experts and the ecological community.
Download my CV here. Last updated April 2024.
Bio for invited talks
Dr. Sara Beery is the Homer A. Burnell Career Development Professor in the MIT Faculty of Artificial Intelligence and Decision-Making. She was previously a visiting researcher at Google, working on automated, large-scale urban forest monitoring and wildlife and plant identification. She received her PhD in Computing and Mathematical Sciences at Caltech, where she was advised by Pietro Perona and awarded the Amori Doctoral Prize for her thesis. Her research focuses on building computer vision methods that enable global-scale environmental and biodiversity monitoring across data modalities, tackling real-world challenges including geospatial and temporal domain shift, learning from imperfect data, fine-grained categories, and long-tailed distributions. She partners with industry, nongovernmental organizations, and government agencies to deploy her methods in the wild worldwide. She works to increase access to AI skills through interdisciplinary capacity building and education, and has founded the AI for Conservation slack community, serves as the Biodiversity Community Lead for Climate Change AI, founded and directs the Workshop on Computer Vision Methods for Ecology, and co-leads the NSF/NSERC Global Center on AI and Biodiversity Change.
In Fall of 2019 I started a Slack channel on AI for Conservation, to provide a shared, interdisciplinary space for researchers who work across the fields of computer vision, machine learning, and AI for conservation and sustainability applications to share opportunities, discuss best practices, and find collaborators. Now our community is over 2500 strong, with researchers from all over the globe. If you'd like to join us, just email (aiforconservation@gmail.com)
