I am a Senior Lecturer and co-lead of the AI group in the School of Computing and Information Systems at the University of Melbourne. Previously, I was a VISTA postdoctoral fellow in the lab of James Elder at the Centre for Vision Research at York University, and a postdoctoral fellow in the lab of Jeremy Wolfe at Brigham & Women's Hospital, Harvard Medical School. I did my PhD at MIT under the supervision of Ruth Rosenholtz.
My work focusses on the intersection of human and computer vision for tasks such as scene recognition, visual search and depth perception in natural scenes. I am interested in developing computer vision algorithms which can visually interpret scenes for place recognition and navigation, and use scene context to support object detection and recognition. I am also interested in how these processes occur in the human visual system. My work combines computational modeling, including Bayesian models and deep neural networks, with behavioral methods, including psychophysics, eye tracking, and large-scale online experiments.
If you are interested in doing a Masters project or PhD in my lab, please follow the link below for more information.
State-Based Disassembly Planning. Lei, C., Lipovetzky, N., & Ehinger, K. A. AAAI Conference on Artificial Intelligence. 2025.
TCAM-Diff: Triplane-Aware Cross-Attention Medical Diffusion Model. Zhang, Z., Ehinger, K. A., & Drummond, T. AAAI Conference on Artificial Intelligence. 2025.
Perceiving Longer Sequences With Bi-Directional Cross-Attention Transformers. Hiller, M., Ehinger, K. A., & Drummond, T. Neural Information Processing Systems (NeurIPS). 2024. pdf
Designing an Adaptive AI System for Operation on Board the SpIRIT Nano-satellite. Joukhadar, Z., Morgan, J., Bayliss, C., Ortiz del Castillo, M., McRobbie, J., Mearns, R., Ehinger, K. A., Rubinstein, B. I. P., Sinnott, R. O., Trenti, M., & Bailey, J. Australasian Joint Conference on Artificial Intelligence (AJCAI). 2024.
End-to-end Truck Speed Detection using Deep Multi-Task Learning. Zuo, H., Sinnott, R. O., & Ehinger, K. A. Australasian Joint Conference on Artificial Intelligence (AJCAI). 2024.
Sequential Amodal Segmentation via Cumulative Occlusion Learning. Ao, J., Ke, Q., & Ehinger, K. A. 35th British Machine Vision Conference (BMVC). 2024.
KALE: An Artwork Image Captioning System Augmented with Heterogeneous Graph. Jiang, Y., Ehinger, K. A., & Lau, J. H. Proceedings of the 33rd International Joint Conference on Artificial Intelligence (IJCAI). 2024.
Mitigating Challenges of the Space Environment for Onboard Artificial Intelligence: Design Overview of the Imaging Payload on SpIRIT. Ortiz del Castillo, M., Morgan, J., McRobbie, J., Therakam, C., Joukhadar, Z., Mearns, R., Barraclough, S.,Sinnott, R., Woods, A., Bayliss, C., Ehinger, K., Rubinstein, B., Bailey, J., Chapman, A., & Trenti, M. Computer Vision and Pattern Recognition (CVPR) Workshop AI4Space 2024: 3rd Workshop on AI for Space. 2024. pdf
Lightness constancy in reality, in virtual reality, and on flat-panel displays. Patel, K. Y., Wilcox, L. M., Maloney, L. T., Ehinger, K. A., Patel, J. Y., Wiedenmann, E., & Murray, R. F. Behavior Research Methods. 2024. pdf
Generalized Planning for the Abstraction and Reasoning Corpus. Lei, C., Lipovetzky, N., & Ehinger, K. A. AAAI Conference on Artificial Intelligence. 2024. pdf
Amodal intra-class instance segmentation: Synthetic datasets and benchmark. Ao, J., Ke, Q., & Ehinger, K. A. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 281-290. 2024. pdf
Exploring automated data augmentation approaches for deep learning: A case study of individual feral cat classification. Yang, Z., Sinnott, R. O., Bailey, J., & Ehinger, K. A. Proceedings of the 57th Hawaii International Conference on System Sciences, pp. 1159-1168. 2024. pdf
Improving denoising diffusion models via simultaneous estimation of image and noise. Zhang, Z., Ehinger, K. A., & Drummond, T. Proceedings of Machine Learning Research 222. 2023. pdf
Truck speed detection through video streams. Huang, Z., Sinnott, R. O., & Ehinger, K. A. Proceedings 2023 19th International Conference on e-Science. 2023. pdf
An active foveated gaze prediction algorithm based on a Bayesian ideal observer. Rashidi, S., Xu, W., Lin, D., Turpin, A., Kulik, L., & Ehinger, K. A. Pattern Recognition. 2023. pdf
Novelty and lifted helpful actions in generalized planning. Lei, C., Lipovetzky, N., & Ehinger, K. A. 16th International Symposium on Combinatorial Search (SoCS). 2023. pdf
Unicode Analogies: An anti-objectivist visual reasoning challenge Spratley, S., Ehinger, K. A., & Miller, T. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 19082-19091. 2023. pdf
Image amodal completion: A survey. Ao, J., Ke, Q., & Ehinger, K. A. Computer Vision and Image Understanding, 229, 103661. 2023. pdf
Hiding the rabbit: Using a genetic algorithm to investigate shape guidance in visual search. Aizenman, A. M., Ehinger, K. A., Wick, F. A., Micheletto, R., Park, J., Jurgensen, L., & Wolfe, J. M. Journal of Vision, 22(1), 7. 2022. pdf
Category systems for real-world scenes. Anderson, M., Graf, E. W., Elder, J. H., Ehinger, K. A., & Adams, W. J. Journal of Vision, 21(2), 8. 2021. pdf
Invertible concept-based explanations for CNN models with non-negative concept activation vectors. Zhang, R., Madumal, P., Miller, T., Ehinger, K. A., & Rubinstein, B. I. P. AAAI Conference on Artificial Intelligence. 2021. pdf
Optimal visual search based on a model of target detectability in natural images. Rashidi, S., Ehinger, K. A., Turpin, A., & Kulik, L. Neural Information Processing Systems (NeurIPS). 2020. pdf
A closer look at generalisation in RAVEN. Spratley, S., Ehinger, K. A., & Miller, T. European Conference on Computer Vision (ECCV). 2020. pdf
Hypergraph optimization for salient region detection based on foreground and background queries. Zhang, J., Fang, S., Ehinger, K. A., Haikun, W., Yang, W., Zhang, K., & Yang, J. IEEE Access, 6, 26729-26741. 2018. pdf
Local depth edge detection in humans and deep neural networks. Ehinger, K. A., Adams, W. J., Graf, E. W., & Elder, J. H. International Conference on Computer Vision (ICCV) Workshop on Mutual Benefits of Cognitive and Computer Vision, 2681-2689. 2017. pdf
Comparing search patterns in digital breast tomosynthesis and full-field digital mammography: an eye tracking study. Aizenman, A. M., Drew, T., Ehinger, K. A., Georgian-Smith, D., & Wolfe, J. M. Journal of Medical Imaging, 4(4):045501, doi: 10.1117/1.JMI.4.4.045501. 2017.
A general account of peripheral encoding also predicts scene perception performance. Ehinger, K. A. & Rosenholtz, R. Journal of Vision, 16:13, doi:10/1167/16.2.13. 2017. pdf
A novel graph-based optimization framework for salient object detection. Zhang, J., Ehinger, K. A., Wei, H., Zhang, K., & Yang, J. Pattern Recognition, 64(C), 39-50. 2017.
When is it time to move to the next map? Optimal foraging in guided search. Ehinger, K. A. & Wolfe, J. M. Attention, Perception, & Psychophysics, 78(7), 2135-2151. 2016. pdf
Change blindness for cast shadows in natural scenes: Even informative shadow changes are missed. Ehinger, K. A., Allen, K., & Wolfe, J. M. Attention, Perception, & Psychophysics, 78(4), 978-987. 2016. website pdf
SUN Database: Exploring a large collection of scene categories. Xiao, J., Ehinger, K. A., Hays, J., Torralba, A., & Oliva, A. International Journal of Computer Vision, 119(1), 3-22. 2016. website pdf
A Change Detection Database for objects in natural indoor scenes. Sareen, P., Ehinger, K. A., & Wolfe, J. M. Behavior Research Methods, 48(4), 1343-1348. 2016. website pdf
TurkerGaze: Crowdsourcing saliency with webcam based eye tracking. Xu, P., Ehinger, K. A, Zhang, Y., Finkelstein, A., Kulkarni, S. R., & Jianxiong, X. arXiv:1504.06755. 2015. website pdf
Through the looking-glass: Objects in the mirror are less real. Sareen, P., Ehinger, K. A., & Wolfe, J. M. Psychonomic Bulletin & Review, 22(4), 980-986. 2015. website pdf
A prior-based graph for salient object detection. Zhang, J., Ehinger, K. A., Ding, J., & Yang, J. Proc. 21st IEEE International Conference on Image Processing (ICIP), 1175 - 1178. 2014. pdf
Basic level scene understanding: Categories, attributes and structures. Xiao, J., Hays, J., Russell, B. C, Patterson, G., Ehinger, K. A., Torralba, A., & Oliva, A. Frontiers in Psychology, 4. doi: 10.3389/fpsyg.2013.00506. 2013. pdf
Basic level scene understanding: From labels to structure and beyond. Xiao, J., Russell, B. C., Hays, J., Ehinger, K. A., Oliva, A., & Torralba, A.. In SIGGRAPH Asia 2012 Technical Briefs (SA '12), Article 36. ACM: New York, NY. 2012 pdf
Recognizing scene viewpoint using panoramic place representation. Xiao, J., Ehinger, K. A., Oliva, A., & Torralba, A. Proc. 25th IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2012. website pdf
How visual and semantic information influence learning in familiar contexts. Goujon, A., Brockmole, J. R., & Ehinger, K. A. Journal of Experimental Psychology: Human Perception and Performance. 2012. pdf
Rethinking the role of top-down attention in vision: Effects attributable to a lossy representation in peripheral vision. Rosenholtz, R., Huang, J., & Ehinger, K. Frontiers in Psychology, 3. doi: 10.3389/fpsyg.2012.00013. 2012. pdf
Canonical views of scenes depend on the shape of the space. Ehinger, K. A., & Oliva, A. In L. Carlson, C. Hölscher, & T. Shipley (Eds.), Proceedings of the 33rd Annual Conference of the Cognitive Science Society (pp. 2114-2119). Austin, TX: Cognitive Science Society. 2011. pdf poster
Estimating scene typicality from human ratings and image features. Ehinger, K. A., Xiao, J., Torralba, A., & Oliva, A. In L. Carlson, C. Hölscher, & T. Shipley (Eds.), Proceedings of the 33rd Annual Conference of the Cognitive Science Society (pp. 2562-2567). Austin, TX: Cognitive Science Society. 2011.pdf slides
What did the early American presidents really look like? Gilbert Stuart portraits as a "Rosetta Stone" to the pre-photography era. Ehinger, K. A., & Altschuler, E. L. Perception, 40(1), 91-94. 2011. website pdf
SUN Database: Large scale scene recognition from abbey to zoo. Xiao, J., Hays, J., Ehinger, K. A., Oliva, A., & Torralba, A. In Proc. 23rd IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 3485-3492. 2010. website pdf
Learning to predict where people look. Judd, T., Ehinger, K., Durand, F., Torralba, A. In 12th IEEE International Conference on Computer Vision (ICCV), 2106-2113. 2009. website pdf
Modeling search for people in 900 Scenes: A combined source model of eye guidance. Ehinger, K. A., Hidalgo-Sotelo, B., Torralba, A., & Oliva, A. Visual Cognition, 17, 945-978. 2009. website pdf
The role of color in visual search in real-world scenes: Evidence from contextual cueing. Ehinger, K. A. & Brockmole, J. R. Perception & Psychophysics, 70(7), 1366-1378. 2008. pdf
Lightness constancy can be very weak in an immersive VR environment. Patel, K. Y., Wilcox, L. M., Maloney, L. T., Ehinger, K. A., Patel, J. Y., & Murray, R. F. Talk presented at Vision Sciences Society annual meeting, 2024.
Role of memory in a Bayesian ideal observer model of visual search in natural images. Rashidi, S., Ehinger, K. A., Kulik, L., & Turpin, A. Talk presented at Vision Sciences Society annual meeting, 2021.
Lightness constancy in reality, in virtual reality, and on flat-panel displays. Patel, K. Y., Wilcox, L. M., Maloney, L. T., Ehinger, K. A., Patel, J. Y., Wiedenmann, E., & Murray, R. F. Poster presented at Vision Sciences Society annual meeting, 2022.
Role of memory in a Bayesian ideal observer model of visual search in natural images. Rashidi, S., Ehinger, K. A., Kulik, L., & Turpin, A. Talk presented at Vision Sciences Society annual meeting, 2021.
Influence of 2D Shape on Contour Depth Perception. Ehinger, K. A., Qian, Y., Wilcox, L. M., & Elder, J. H. Talk presented at Vision Sciences Society annual meeting, May 18, 2019. slides
Use of local image information in depth edge classification by humans and neural networks. Ehinger, K. A., Adams, W. J., Graf, E. W., & Elder, J. H. Talk presented at MODVIS (Computational and Mathematical Models in Vision), May 17, 2018.
Learning to identify depth edges in real-world images with 3D ground truth. Ehinger, K. A., Joseph, K. T., Adams, W. J., Graf, E. W., & Elder, J. H. Talk presented at MODVIS (Computational and Mathematical Models in Vision), May 19, 2017.
Foraging in satellite imagery: When is it time to move to the next map? Ehinger, K. A. & Wolfe, J. M. Talk presented at Vision Sciences Society annual meeting, May 18, 2015.
Foraging and navigating in a virtual orchard: Which tree do you visit next? Ehinger, K. A. & Wolfe, J. M. Talk presented at Vision Sciences Society annual meeting, May 19, 2014.
What determines the canonical view of a scene? Ehinger, K. A. & Oliva, A. Talk presented at Vision Sciences Society annual meeting, May 8, 2011.
Canonical views of scenes. Ehinger, K. A. & Oliva, A. Poster presented at the MIT Scene Understanding Symposium, Jan 28, 2011.
Modeling search for people in 900 Scenes: The roles of saliency, target features, and scene context. Ehinger, K. A., Hidalgo-Sotelo, B., Torralba, A., & Oliva, A. Talk presented at Vision Sciences Society annual meeting, May 10, 2009. slides
Modeling sources of visual attention guidance in real-world search. Ehinger, K. A. Talk presented at MIT Cognitive Lunch, Oct 21, 2008.
Stimuli images, fixation data, and analysis code from Modeling search for people in 900 Scenes.
Collection of over 200 natural scene pairs with object changes for use in change blindness experiments.
Scene recognition database consisting of 899 scene categories, object/region annotations.
Panoramic scene dataset consisting of 369 scene categories, with MATLAB analysis code.
Stimuli images, human fixation dataset, and saliency model from Learning to predict where humans look.
Sample code for running various types of visual cognition experiments in MATLAB with PsychToolbox.
A simple implementation of Hays & Efros' Scene Completion Using Millions of Photographs in MATLAB.
