About

Andrew Owens is an associate professor of computer science at Cornell Tech and the Cornell Ann S. Bowers College of Computing and Information Science. His research aims to create multimodal systems that learn to see, hear, and touch without human-labeled training data. Instead, these systems learn from co-occurring sensory signals, such as the correlations between the visual and audio streams of a video. His work has enabled applications that include producing soundtracks for silent videos, robotic manipulation with vision and touch, detecting AI-generated images, and generating visual illusions. Owens is a recipient of a Sloan Research Fellowship and an NSF CAREER Award. Prior to joining Cornell, he was an assistant professor at the University of Michigan and a postdoctoral scholar at the University of California, Berkeley. He received a Ph.D. in computer science from the Massachusetts Institute of Technology in 2016 and a B.A. in computer science from Cornell University in 2010.

Research topics

• Computer Science
• Artificial Intelligence
• Computer vision

Selected publications

• Investigating Screen Complexity in the Vehicle, How Does Design Effect Driver Attention and Task Performance?

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• GPS as a Control Signal for Image Generation

  2025-06-10 · 2 citations

  articleSenior author

  We show that the GPS tags contained in photo metadata provide a useful control signal for image generation. We train GPS-to-image models and use them for tasks that require a fine-grained understanding of how images vary within a city. In particular, we train a diffusion model to generate images conditioned on both GPS and text. The learned model generates images that capture the distinctive appearance of different neighborhoods, parks, and landmarks. We also extract 3D models from 2D GPS-to-image models through score distillation sampling, using GPS conditioning to constrain the appearance of the reconstruction from each viewpoint. Our evaluations suggest that our GPS-conditioned models successfully learn to generate images that vary based on location, and that GPS conditioning improves estimated 3D structure.

  PublisherDOI

• Contrastive Touch-to-Touch Pretraining

  2025-05-19 · 1 citations

  article

  Today's tactile sensors have a variety of different designs, making it challenging to develop general-purpose methods for processing touch signals. In this paper, we learn a unified representation that captures the shared information between different tactile sensors. Unlike current approaches that focus on reconstruction or task-specific supervision, we leverage contrastive learning to integrate tactile signals from two different sensors into a shared embedding space, using a dataset in which the same objects are probed with multiple sensors. We apply this approach to paired touch signals from GelSlim and Soft Bubble sensors. We show that our learned features provide strong pretraining for downstream pose estimation and classification tasks. We also show that our embedding enables models trained using one touch sensor to be deployed using another without additional training. Project details can be found at https://www.mmintlab.com/research/cttp/.

  PublisherDOI

• Community Forensics: Using Thousands of Generators to Train Fake Image Detectors

  2025-06-10 · 3 citations

  articleSenior author

  One of the key challenges of detecting AI-generated images is spotting images that have been created by previously unseen generative models. We argue that the limited diversity of the training data is a major obstacle to addressing this problem, and we propose a new dataset that is significantly larger and more diverse than prior works. As part of creating this dataset, we systematically download thousands of text-to-image latent diffusion models and sample images from them. We also collect images from dozens of popular open source and commercial models. The resulting dataset contains 2.7M images that have been sampled from 4803 different models. These images collectively capture a wide range of scene content, generator architectures, and image processing settings. Using this dataset, we study the generalization abilities of fake image detectors. Our experiments suggest that detection performance improves as the number of models in the training set increases, even when these models have similar architectures. We also find that increasing the diversity of the models improves detection performance, and that our trained detectors generalize better than those trained on other datasets. The dataset can be found in https://jespark.net/projects/2024/community_forensics

  PublisherDOI

• Supervising Sound Localization by In-the-wild Egomotion

  2025-06-10

  articleSenior author

  We present a method for learning binaural sound localization using egomotion as a supervisory signal. Over the course of a video, the camera’s direction to a sound source will change as the camera moves. We train an audio model to predict sound directions that are consistent with visual estimates of camera motion, which we obtain using traditional methods from multi-view geometry. This provides a weak but plentiful form of supervision that we combine with traditional binaural cues. To evaluate this method, we propose a dataset of real-world audio-visual videos with egomotion. We show that our model can successfully learn from real-world data and that it performs well on sound localization tasks.

  PublisherDOI

• Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models

  2024-06-16 · 13 citations

  articleSenior author

  We address the problem of synthesizing multi-view optical illusions: images that change appearance upon a transformation, such as a flip or rotation. We propose a simple, zero-shot method for obtaining these illusions from off-the-shelf text-to-image diffusion models. During the reverse diffusion process, we estimate the noise from different views of a noisy image, and then combine these noise estimates together and denoise the image. A theoretical analysis suggests that this method works precisely for views that can be written as orthogonal transformations, of which permutations are a subset. This leads to the idea of a visual anagram-an image that changes appearance under some rearrangement of pixels. This includes rotations and flips, but also more exotic pixel permutations such as a jigsaw rearrangement. Our approach also naturally extends to illusions with more than two views. We provide both qualitative and quantitative results demonstrating the effectiveness and flexibility of our method. Please see our project webpage for additional visualizations and results: https://dangeng.github.io/visual_anagrams/.

  PublisherDOI

• Efficient Vision-Language Pre-Training by Cluster Masking

  2024-06-16 · 4 citations

  articleSenior author

  We propose a simple strategy for masking image patches during visual-language contrastive learning that improves the quality of the learned representations and the training speed. During each iteration of training, we randomly mask clusters of visually similar image patches, as measured by their raw pixel intensities. This provides an extra learning signal, beyond the contrastive training itself, since it forces a model to predict words for masked visual structures solely from context. It also speeds up training by reducing the amount of data used in each image. We evaluate the effectiveness of our model by pre-training on a number of bench-marks, finding that it outperforms other masking strategies, such as FLIP, on the quality of the learned representation.

  PublisherDOI

• MaPP Puzzle Hunt

  Math Horizons · 2024-08-29

  articleSenior author
  PublisherDOI

• Community Forensics: Using Thousands of Generators to Train Fake Image Detectors

  arXiv (Cornell University) · 2024-11-06

  preprintOpen accessSenior author

  One of the key challenges of detecting AI-generated images is spotting images that have been created by previously unseen generative models. We argue that the limited diversity of the training data is a major obstacle to addressing this problem, and we propose a new dataset that is significantly larger and more diverse than prior work. As part of creating this dataset, we systematically download thousands of text-to-image latent diffusion models and sample images from them. We also collect images from dozens of popular open source and commercial models. The resulting dataset contains 2.7M images that have been sampled from 4803 different models. These images collectively capture a wide range of scene content, generator architectures, and image processing settings. Using this dataset, we study the generalization abilities of fake image detectors. Our experiments suggest that detection performance improves as the number of models in the training set increases, even when these models have similar architectures. We also find that detection performance improves as the diversity of the models increases, and that our trained detectors generalize better than those trained on other datasets. The dataset can be found in https://jespark.net/projects/2024/community_forensics

  PublisherOA PDFDOI

• Contrastive Touch-to-Touch Pretraining

  arXiv (Cornell University) · 2024-10-15

  preprintOpen access

  Today's tactile sensors have a variety of different designs, making it challenging to develop general-purpose methods for processing touch signals. In this paper, we learn a unified representation that captures the shared information between different tactile sensors. Unlike current approaches that focus on reconstruction or task-specific supervision, we leverage contrastive learning to integrate tactile signals from two different sensors into a shared embedding space, using a dataset in which the same objects are probed with multiple sensors. We apply this approach to paired touch signals from GelSlim and Soft Bubble sensors. We show that our learned features provide strong pretraining for downstream pose estimation and classification tasks. We also show that our embedding enables models trained using one touch sensor to be deployed using another without additional training. Project details can be found at https://www.mmintlab.com/research/cttp/.

  PublisherOA PDFDOI

Frequent coauthors

• Alexei A. Efros

  34 shared

• Sheng-Yu Wang

  Carnegie Mellon University

  13 shared

• William T. Freeman

  11 shared

• Shiry Ginosar

  10 shared

• Przemysław Prusinkiewicz

  University of Calgary

  8 shared

• Mikolaj Cieslak

  University of Calgary

  8 shared

• Oliver Wang

  Adobe Systems (United States)

  8 shared

• Richard Zhang

  8 shared

Awards & honors

• Sloan Research Fellowship
• NSF CAREER Award