Menu
Explore a research direction
Computer vision
research labs.
Discover research in computer vision, image processing, and visual computing through faculty-led lab profiles. Review publications and official sources to understand each group's specific focus.
What to compare
Compare the kinds of visual data and problems a group studies. Look for differences between image analysis, video, three-dimensional scenes, and domain-specific applications, and ask how results are evaluated.
Profiles appear here because their listed research keywords match this topic. Inclusion does not establish an open position. Check the official sources and any published application instructions before reaching out.
Prepare a research introduction ↗45 research-led lab profiles
Browse labs by their listed college affiliationAlexander Schwing Research Group
University of Illinois at Urbana-ChampaignAlexander Schwing · Associate Professor
How can machines learn to see and understand images, a question studied in machine learning and computer vision? Alexander Schwing is an Associate Professor in Electrical and Computer Engineering at the University of Illinois. His page lists a faculty fellowship supported by Advanced Micro Devices, Inc. His listed research areas include computer vision and pattern recognition, machine learning, and robotics, vision, and artificial intelligence. He teaches a machine learning course, CS 446, also listed as ECE 449.
Alexei A. Efros Research Group
University of California, BerkeleyAlexei A. Efros · Howard Friesen Professor
How can vast amounts of unlabelled visual data be used to understand, model, and recreate the visual world around us? Alexei Efros works mainly in data-driven computer vision, together with its projection onto computer graphics and computational photography. He is part of the Berkeley Artificial Intelligence Research Lab (BAIR) in the EECS Department at UC Berkeley. His page describes work on autoregressive image generation since 1999 and visual data scaling since 2007.
Ali Farhadi Research Group
University of WashingtonAli Farhadi · Professor
How can computer vision and natural language be connected? Ali Farhadi is a Professor in the Department of Computer Science and Engineering at the University of Washington. His interests are mainly computer vision and machine learning, including the intersection of natural language and vision. He also studies the role of semantics in visual understanding and visual reasoning. His homepage points to the RAIVN Lab for more recent updates on his work.
Arie Kaufman Research Group
Stony Brook UniversityArie Kaufman · Distinguished Professor
Complex data become easier to investigate when researchers can see and interact with their structure. Kaufman works in computer graphics and visualization, including methods for displaying volumetric and scientific data. His interests extend to interfaces, virtual reality, multimedia, and medical imaging. Current university research profiles highlight work combining virtual reality and machine learning with three dimensional virtual pancreatography for pancreatic cancer research. As director of the Center of Visual Computing, he connects visual computing methods with applications that help researchers examine and interpret complex information.
Arun Ross Research Group
Michigan State UniversityArun Ross · Martin J. Vanderploeg Endowed Professor, Computer Science and Engineering
How can biometric systems be made resilient to presentation attacks while preserving user privacy and recognition accuracy? The group researches presentation attack detection methods and spoofing defenses, including explainable attention-guided iris PAD. Researchers develop semi-adversarial networks and convolutional autoencoders to impart privacy to face images while retaining utility for recognition. The work explores vulnerabilities in fingerprint and iris systems through empirical studies that expose spoofing and masterprint risks. The lab investigates multi-biometrics and fusion strategies to improve recognition reliability under adversarial or degraded inputs.
Baowei Fei Research Group
University of Texas at DallasBaowei Fei · Professor
Can imaging reveal information that a surgeon cannot see with the naked eye? Fei develops medical imaging and computational methods for identifying tissue and guiding interventions. His research includes hyperspectral imaging, which measures light across many wavelengths, paired with artificial intelligence to investigate cancer detection. Other work combines imaging modalities such as PET and three-dimensional ultrasound for more precise biopsy guidance. The group connects engineering algorithms and imaging hardware with clinical questions in areas including head and neck and prostate cancer, making image-guided medicine a central research theme.
Bharath Hariharan Research Group
Cornell UniversityBharath Hariharan · Associate Professor
How can machine learning be combined with computer vision for problems that defy the Big Data label? Bharath Hariharan is an associate professor in Computer Science at Cornell University. He enjoys problems that require marrying advances in machine learning with insights from computer vision, geometry and domain-specific knowledge. His group works on recognition for satellite images and earth science. The group is also exploring 4D reconstruction and recognition, including tracking individual pixels through long-term occlusions.
Bhiksha Raj Research Group
Carnegie Mellon UniversityBhiksha Raj · Professor
How can machine learning advance speech processing and computer vision? The research of Bhiksha Raj spans machine learning, speech processing, computer vision, and time-series analysis. Applications range from audio analysis to deep learning and quantum computing. He has contributed to robust speech recognition, privacy-preserving machine learning, and neural network architectures. Recent work also explores the intersection of large-scale machine learning models and signal processing, with a focus on real-world challenges.
Chenliang Xu Research Group
University of RochesterChenliang Xu · Associate Professor of Computer Science
How can video understanding methods link vision and language to describe actions and narratives in moving images? The group develops methods for video segmentation and fine-grained actor-action segmentation to separate actors and actions across frames using computer vision algorithms. Researchers build activity recognition and video storytelling models that combine visual features with natural language representations to generate coherent descriptions of video content. The group explores cross-modal audio-visual generation to synthesize complementary modalities for richer video understanding using deep multimodal learning. Recent projects include work on multimodal vision-and-x modeling and trustworthy AI for video tasks.
Danfei Xu Research Group
Georgia Institute of TechnologyDanfei Xu · Assistant Professor
How can robots achieve physical autonomy in everyday human environments with minimum expert intervention? Danfei Xu is an Assistant Professor in the School of Interactive Computing at Georgia Tech. His work draws equally from Robotics, Machine Learning, and Computer Vision, including topics such as imitation and reinforcement learning, representation learning, manipulation, and human-robot interaction. His current research focuses on visuomotor skill learning, structured world models for long-horizon planning, and data-driven approaches to human-robot collaboration. The profile lists the research areas Robotics, Artificial Intelligence, Machine Learning, and Computer Vision.
David Alexander Forsyth Research Group
University of Illinois at Urbana-ChampaignDavid Alexander Forsyth · Professor, Siebel School of Computing and Data Science
How can computers be built to see and to learn from images, graphics and other data? David Forsyth's listed research interests are computer vision, computer graphics, and machine learning. His research areas include Artificial Intelligence and Interactive Computing. He has recently taught the course CS 588 on autonomous vehicle system engineering at the university. He has also taught a vision seminar, listed as CS 591 DAF, among his recent courses taught.
Deepak Pathak Research Group
Carnegie Mellon UniversityDeepak Pathak · Associate Professor
How can agents learn to generalize across real and diverse environments, using ideas from computer vision, machine learning and robotics? Deepak Pathak is a Raj Reddy Associate Professor at Carnegie Mellon University in the School of Computer Science. He is a member of the Robotics Institute and is affiliated with the Machine Learning Department. He works in Artificial Intelligence at the intersection of Computer Vision, Machine Learning and Robotics. The research group draws inspiration from psychology to build practical systems at the interface of vision, learning and robotics.
Derek Hoiem Research Group
University of Illinois at Urbana-ChampaignDerek Hoiem · Professor
How can computers be taught to understand images and scenes? Derek Hoiem is a Professor in the Siebel School for Computing and Data Science at the University of Illinois Urbana-Champaign. He earned his PhD in Robotics from Carnegie Mellon University in 2007. His research interests are computer vision, object recognition, scene understanding, and graphics. He is co-founder and Chief Science Officer of Reconstruct, which provides an AI-based Visual Command Center to construction stakeholders.
Deva Ramanan Research Group
Carnegie Mellon UniversityDeva Ramanan · Professor
How can machine learning help computer vision systems make sense of the visual world? The page says that the human visual system is often used as inspiration for the research. It describes temporal processing as a key component of human perception that is still relatively unexploited in current visual recognition systems. It says that machine learning from big visual data allows systems to learn subtle statistical regularities of the visual world. The page also notes that humans have the ability to learn from very few examples, in contrast to current systems.
Dimitris Metaxas Research Group
Rutgers University–New BrunswickDimitris Metaxas · Board of Governors Professor
Metaxas investigates how artificial intelligence can combine learned patterns with knowledge about the world. His research connects machine learning with physical principles, dynamical systems, and other domain knowledge. Projects span computer vision, biomedical image analysis, and computer graphics, alongside image and text generation using generative models. Additional interests include language and vision language models, explainable AI, and learning with limited supervision. These directions connect the mathematical design of intelligent systems with applications where understanding images and interpreting complex data are central research challenges.
Donglai Wei Research Group
Boston CollegeDonglai Wei · Associate Professor
How can computational models for biomedical image analysis, such as connectomics, be built? Donglai Wei's research interests concern the interplay between biological sciences and computer vision. The research also examines brain mechanisms to guide machine video understanding. Selected publications include the MitoEM dataset for large-scale 3D mitochondria instance segmentation from EM images. Other selected publications include video enhancement with task-oriented flow and a microscope for visualizing and quantifying small motions. Another selected publication is on learning and using the arrow of time.
Edward Adelson Research Group
Massachusetts Institute of TechnologyEdward Adelson · John and Dorothy Wilson Professor of Vision Science
How do human vision and computer vision solve problems in visual perception? Edward Adelson is the John and Dorothy Wilson Professor of Vision Science at MIT, in the Department of Brain and Cognitive Sciences. Current topics include mid-level visual processing, including perceptual organization, as applied to motion, transparency, lightness, and texture. The group also studies the perception of materials, such as how it is possible to tell that something is shiny or translucent. His lab is now focused on artificial touch sensing for robotics, using an elastomeric technology to build robot fingers that are soft and sensitive.
Frank Dellaert Research Group
Georgia Institute of TechnologyFrank Dellaert · Professor
How can Bayesian inference solve the difficult inverse problems that arise in robotics and computer vision? Dr. Dellaert does research in the areas of robotics and computer vision. He is interested in whether Monte Carlo sampling-based approximations are applicable when exact solutions are intractable. His additional research includes advanced sequential Monte Carlo methods and simultaneous localization and mapping. He also lists spatio-temporal reconstruction from images among his additional research areas at Georgia Tech.
Irfan A. Essa Research Group
Georgia Institute of TechnologyIrfan A. Essa · Distinguished Professor
Which problems in computer vision, artificial intelligence and machine learning does Irfan Essa work on? Irfan Essa is a Distinguished Professor at the School of Interactive Computing in the College of Computing at Georgia Tech. He is the Executive Director of the Interdisciplinary Research Center for Machine Learning at Georgia Tech (ML@GT), which he founded in 2015. His areas of work include computer graphics, computational perception, robotics, computer animation, and social computing. His work also relates to human-computer interaction and computational behavioral and social sciences.
James Hays Research Group
Georgia Institute of TechnologyJames Hays · Professor
How can machines learn to recognize, synthesize, and manipulate visual content by exploiting newly found or newly created data? The research spans computer vision, robotics, and machine learning, with problems organized around recognition, synthesis, and manipulation as the central themes. Researchers often look for new data sources to exploit, such as geotagged imagery that ties photographs to real places. Where suitable data is missing, the team builds entirely new datasets, for example collections of sketches or of grasps that did not exist before. A further applied project aims to develop off-road autonomous vehicles, working together with an industry partner named Overland AI.
Jiaqi (Jackey) Gong Research Group
The University of AlabamaJiaqi (Jackey) Gong · Associate Professor, Adjunct Associate Professor of ME, Director for the Alabama Center for the Advancement of Artificial Intelligence
Jiaqi (Jackey) Gong's faculty-led research profile covers Artificial Intelligence, Computer Vision, Data Analytics, and Deep Learning.
Judy Hoffman Research Group
Georgia Institute of TechnologyJudy Hoffman · Associate Professor
How can computer vision and machine learning be combined to adapt models across domains? Judy Hoffman is an associate professor in the School of Interactive Computing at Georgia Tech. Her research specializes in domain adaptation, transfer learning, adversarial robustness, and algorithmic fairness. She is a member of the Machine Learning Center. She co-founded Women in Computer Vision, which provides mentorship and travel support for early-career women in the computer vision community.
Judy Hoffman Research Group
University of California, IrvineJudy Hoffman · Associate Professor of Computer Science
How can learning-based AI systems reliably benefit a diverse set of people and applications? Judy Hoffman is an Associate Professor of Computer Science in the Donald Bren School of Information and Computer Sciences at the University of California, Irvine. Her research lies at the intersection of computer vision and machine learning. Her research aims to enable learning based AI systems that can work even under resource constraints. She received her PhD in Electrical Engineering and Computer Science from UC Berkeley.
Jun-Yan Zhu Research Group
Carnegie Mellon UniversityJun-Yan Zhu · Michael B. Donahue Associate Professor of Computer Science and Robotics
How can human creators and generative models collaborate so creators control outputs more easily and tell visual stories? The group develops interfaces and algorithms that let creators steer controllable visual synthesis of images, videos, 3D content, and combined visual and tactile data. Another question is how creators can adapt existing models to new tasks, concepts, and styles, and even rewrite the rules those models follow. Researchers also ask how creators could opt into or out of generative systems whenever they choose, and how contributors of training data should be credited. Generative models are also used to produce synthetic data that improves computer vision and robotics systems.
Showing 1–24 of 45 labs