Oisin Mac Aodha

My name is Oisin Mac Aodha and I am a postdoc in Prof. Pietro Perona's Computational Vision Lab at Caltech. My current research interest is in the areas of computer vision and machine learning, with a specific emphasis on human-in-the-loop methods such as machine teaching.

Previous to this, I spent three great years as a postdoc in the Department of Computer Science at the University College of London with Prof. Gabriel Brostow and Prof. Kate Jones. There I worked on interactive machine learning, where our goal was to design algorithms to enable non-programming scientists to semi-automatically explore events of interest in vast quantities of audio and visual data.

I did both my MSc (with Dr. Simon Prince) and PhD (with Prof. Gabriel Brostow) at UCL and have an undergraduate degree in electronic engineering from NUI Galway in Ireland. Before I started my PhD I was a research assistant for one year in Prof. Marc Pollefeys' group at ETH Zurich.

macaodha (at) caltech.edu

Oisin

News

Jun 2018: FGVC5 held at CVPR 2018.
Feb 2018: Launched iNaturalist 2018 challege.
Jan 2018: Co-organizing FGVC5 workshop at CVPR 2018.
August 2017: Read about the iNaturalist Challenge 2017 that we ran in conjunction with Caltech, iNaturalist, Cornell, Google, and Kaggle.
July 2017: Helped organize the Fourth Fine-Grained Visual Categorization workshop at CPVR this summer.
July 2017: Read about our monocular depth estimation project at UCL and The Engineer.
June 2017: Live deployment of our bat detection system in the former Olympic Park in London. Read about it on the BBC or follow our twitter page to get daily updates.
Sept 2016: Interviewed by SVT for our work on using machine learning to detect bats.
August 2016: Read about our handwriting synthesis project at the BBC, UCL, Engadget, Reuters, Gizmodo, and The Irish Times.
July 2016: Moved to Caltech to do a postdoc.
March 2016: Gave a talk about Bat Detective at UCL's Grant Museum.
April 2015: Blog about bat call detection over at Methods in Ecology and Evolution.

Publications

More details on Google Scholar.

Digging Into Self-Supervised Monocular Depth Estimation
Monodepth 2.0 Clement Godard, Oisin Mac Aodha, and Gabriel Brostow
Preprint 2018
[preprint] [video]
Teaching Multiple Concepts to Forgetful Learners
Forgetful Anette Hunziker, Yuxin Chen, Oisin Mac Aodha, Manuel Gomez Rodriguez, Andreas Krause, Pietro Perona, Yisong Yue, and Adish Singla
Preprint 2018
[preprint]
Understanding the Role of Adaptivity in Machine Teaching: The Case of Version Space Learners
Adaptive 2018 Yuxin Chen, Adish Singla, Oisin Mac Aodha, Pietro Perona, and Yisong Yue
NeurIPS 2018
[paper]
It's all Relative: Monocular 3D Human Pose Estimation from Weakly Supervised Data
Rel Pose Matteo Ruggero Ronchi, Oisin Mac Aodha, Robert Eng, and Pietro Perona
BMVC 2018 (Oral)
[paper] [webpage] [video]
Teaching Categories to Human Learners with Visual Explanations
Teach CVPR 2018 Oisin Mac Aodha, Shihan Su, Yuxin Chen, Pietro Perona, and Yisong Yue
CVPR 2018 (Spotlight)
[paper] [code] [video]
Context Embedding Networks
CEN Kun ho Kim, Oisin Mac Aodha, and Pietro Perona
CVPR 2018 (Spotlight)
[paper]
The iNaturalist Species Classification and Detection Dataset
iNat 2017 Grant Van Horn, Oisin Mac Aodha, Yang Song, Yin Cui, Chen Sun, Alex Shepard, Hartwig Adam, Pietro Perona, and Serge Belongie
CVPR 2018 (Spotlight)
[paper] [webpage] [video]
Bat Detective - Deep Learning Tools for Bat Acoustic Signal Detection
Bat Detective Oisin Mac Aodha, Rory Gibb, Kate Barlow, Ella Browning, Michael Firman, Robin Freeman, Briana Harder, Libby Kinsey, Gary Mead, Stuart Newson, Ivan Pandourski, Stuart Parsons, Jon Russ, Abigel Szodoray-Paradi, Farkas Szodoray-Paradi, Elena Tilova, Mark Girolami, Gabriel J. Brostow, and Kate E. Jones
PLOS Computational Biology 2018 [paper] [webpage]
Near-Optimal Machine Teaching via Explanatory Teaching Sets
AIStats 2018 Yuxin Chen, Oisin Mac Aodha, Shihan Su, Pietro Perona, and Yisong Yue
AISTATS 2018
[paper]
Interpretable Machine Teaching via Feature Feedback
NIPS WS 2017 Shihan Su, Yuxin Chen, Oisin Mac Aodha, Pietro Perona, and Yisong Yue
NeurIPS Workshop Teaching Machines, Robots, and Humans 2017
[paper]
Unsupervised Monocular Depth Estimation with Left-Right Consistency
CVPR 2017 Clement Godard, Oisin Mac Aodha, and Gabriel J. Brostow
CVPR 2017 (Oral)
[paper] [webpage]
Structured Prediction of Unobserved Voxels From a Single Depth Image
CVPR 2016 Michael Firman, Oisin Mac Aodha, Simon Julier, and Gabriel J. Brostow
CVPR 2016 (Oral)
[paper] [webpage]
My text in Your Handwriting
ToG 2016 Tom S.F. Haines, Oisin Mac Aodha, and Gabriel J. Brostow
Transactions on Graphics 2016
[paper] [webpage]
Becoming the Expert - Interactive Multi-Class Machine Teaching
CVPR 2015 Edward Johns, Oisin Mac Aodha, and Gabriel J. Brostow
CVPR 2015 (Oral)
[paper] [webpage]
Putting the Scientist in the Loop - Accelerating Scientific Progress with Interactive Machine Learning
ICPR 2014 Oisin Mac Aodha, Vassilios Stathopoulos, Michael Terry, Kate E. Jones, Gabriel J. Brostow, and Mark Girolami
ICPR 2014
[paper]
Hierarchical Subquery Evaluation for Active Learning on a Graph
CVPR 2014 Oisin Mac Aodha, Neill D.F. Campbell, Jan Kautz, and Gabriel J. Brostow
CVPR 2014 (Oral)
[paper] [webpage]
Revisiting Example Dependent Cost-Sensitive Learning with Decision Trees
ICCV 2013 Oisin Mac Aodha and Gabriel J. Brostow
ICCV 2013
[paper] [webpage]
Learning a Confidence Measure for Optical Flow
PAMI 2013 Oisin Mac Aodha, Ahmad Humayun, Marc Pollefeys, and Gabriel J. Brostow
PAMI 2013
[paper] [webpage]
Patch Based Synthesis for Single Depth Image Super-Resolution
ECCV 2012 Oisin Mac Aodha, Neill D.F. Campbell, Arun Nair, and Gabriel J. Brostow
ECCV 2012
[paper] [webpage]
Learning to Find Occlusion Regions
CVPR 2011 Ahmad Humayun, Oisin Mac Aodha, and Gabriel J. Brostow
CVPR 2011
[paper] [webpage]
Segmenting Video Into Classes of Algorithm-Suitability
CVPR 2010 Oisin Mac Aodha, Gabriel J Brostow, and Marc Pollefeys
CVPR 2010
[paper] [webpage]
Evolving Plastic Responses in Artificial Cell Models
Cell Im John Maher, Fearghal Morgan, and Oisin Mac Aodha
Congress on Evolutionary Computation 2009
[paper]

Theses

Supervised Algorithm Selection for Flow and Other Computer Vision Problems
PhD PhD Thesis
UCL 2014
[thesis]
Reconstruction and Recognition of 3D Face Models
3DFace MSc Thesis
UCL 2008
[thesis]