[ { "id": "https://authors.library.caltech.edu/records/pvmcz-13r80", "eprint_id": 121550, "eprint_status": "archive", "datestamp": "2023-08-20 16:54:08", "lastmod": "2023-10-23 15:39:15", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Gao-Angela-F", "name": { "family": "Gao", "given": "Angela F." }, "orcid": "0000-0001-8574-8728" }, { "id": "Leong-Oscar", "name": { "family": "Leong", "given": "Oscar" } }, { "id": "Sun-He", "name": { "family": "Sun", "given": "He" }, "orcid": "0000-0003-1526-6787" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" } ] }, "title": "Image Reconstruction without Explicit Priors", "ispublished": "unpub", "full_text_status": "public", "note": "\u00a9 2023 IEEE. \n\nThis work was sponsored by NSF Award 2048237 and 1935980, an Amazon AI4Science Partnership Discovery Grant, and the Caltech/JPL President's and Director's Research and Development Fund (PDRDF). This research was carried out at the Jet Propulsion Laboratory and Caltech under a contract with the National Aeronautics and Space Administration and funded through the PDRDF.", "abstract": "We consider solving ill-posed imaging inverse problems without access to an explicit image prior or ground-truth examples. An overarching challenge in inverse problems is that there are many undesired images that fit to the observed measurements, thus requiring image priors to constrain the space of possible solutions to more plausible reconstructions. However, in many applications it is difficult or potentially impossible to obtain ground-truth images to learn an image prior. Thus, inaccurate priors are often used, which inevitably result in biased solutions. Rather than solving an inverse problem using priors that encode the explicit structure of any one image, we propose to solve a set of inverse problems jointly by incorporating prior constraints on the collective structure of the underlying images. The key assumption of our work is that the ground-truth images we aim to reconstruct share common, low-dimensional structure. We show that such a set of inverse problems can be solved simultaneously by learning a shared image generator with a low-dimensional latent space. The parameters of the generator and latent embedding are learned by maximizing a proxy for the Evidence Lower Bound (ELBO). Once learned, the generator and latent embeddings can be combined to provide reconstructions for each inverse problem. The framework we propose can handle general forward model corruptions, and we show that measurements derived from only a few ground-truth images (O(10)) are sufficient for image reconstruction without explicit priors.", "date": "2023-06", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "id_number": "CaltechAUTHORS:20230526-662984000.7", "isbn": "9781728163277", "book_title": "2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20230526-662984000.7", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "CCF-2048237" }, { "agency": "NSF", "grant_number": "AST-1935980" }, { "agency": "Amazon AI4Science Fellowship" }, { "agency": "JPL President and Director's Fund" }, { "agency": "NASA/JPL/Caltech" } ] }, "doi": "10.1109/icassp49357.2023.10096515", "resource_type": "book_section", "pub_year": "2023", "author_list": "Gao, Angela F.; Leong, Oscar; et el." }, { "id": "https://authors.library.caltech.edu/records/jmn81-k7q70", "eprint_id": 117607, "eprint_status": "archive", "datestamp": "2023-08-20 06:09:18", "lastmod": "2023-10-24 22:35:44", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Gao-Angela-F", "name": { "family": "Gao", "given": "Angela F." }, "orcid": "0000-0001-8574-8728" }, { "id": "Castellanos-Jorge-C", "name": { "family": "Castellanos", "given": "Jorge C." }, "orcid": "0000-0002-0103-6430" }, { "id": "Yue-Yisong", "name": { "family": "Yue", "given": "Yisong" }, "orcid": "0000-0001-9127-1989" }, { "id": "Ross-Z-E", "name": { "family": "Ross", "given": "Zachary E." }, "orcid": "0000-0002-6343-8400" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" } ] }, "title": "DeepGEM: Generalized Expectation-Maximization for Blind Inversion", "ispublished": "unpub", "full_text_status": "public", "note": "This research was carried out at the Jet Propulsion Laboratory and the California Institute of Technology under a contract with the National Aeronautics and Space Administration and funded through the President's and Director's Research and Development Fund (PDRDF). This work was sponsored by Beyond Limits, Jet Propulsion Laboratory Award 1669417, NSF Award 2048237, and generous gifts from Luke Wang and Yi Li. Additionally, we would like to thank He Sun for many helpful discussions. We declare no competing interests.\n\n
Published - GAO_ANIPS_2021.pdf
", "abstract": "Typically, inversion algorithms assume that a forward model, which relates a source to its resulting measurements, is known and fixed. Using collected indirect measurements and the forward model, the goal becomes to recover the source. When the forward model is unknown, or imperfect, artifacts due to model mismatch occur in the recovery of the source. In this paper, we study the problem of blind inversion: solving an inverse problem with unknown or imperfect knowledge of the forward model parameters. We propose DeepGEM, a variational Expectation-Maximization (EM) framework that can be used to solve for the unknown parameters of the forward model in an unsupervised manner. DeepGEM makes use of a normalizing flow generative network to efficiently capture complex posterior distributions, which leads to more accurate evaluation of the source's posterior distribution used in EM. We showcase the effectiveness of our DeepGEM approach by achieving strong performance on the challenging problem of blind seismic tomography, where we significantly outperform the standard method used in seismology. We also demonstrate the generality of DeepGEM by applying it to a simple case of blind deconvolution.", "date": "2021-12", "date_type": "published", "publisher": "Neural Information Processing Systems foundation", "place_of_pub": "La Jolla, CA", "pagerange": "1-12", "id_number": "CaltechAUTHORS:20221026-200931912", "isbn": "9781713845393", "book_title": "35th Conference on Neural Information Processing Systems", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20221026-200931912", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NASA/JPL/Caltech" }, { "agency": "JPL President and Director's Fund" }, { "agency": "Beyond Limits" }, { "agency": "JPL", "grant_number": "1669417" }, { "agency": "NSF", "grant_number": "CCF-2048237" } ] }, "local_group": { "items": [ { "id": "Seismological-Laboratory" }, { "id": "Astronomy-Department" }, { "id": "Center-for-Geomechanics-and-Mitigation-of-Geohazards-(GMG)" }, { "id": "Division-of-Geological-and-Planetary-Sciences" } ] }, "contributors": { "items": [ { "id": "Ranzato-M", "name": { "family": "Ranzato", "given": "M." } }, { "id": "Beygelzimer-A", "name": { "family": "Beygelzimer", "given": "A." } }, { "id": "Dauphin-Y", "name": { "family": "Dauphin", "given": "Y." } }, { "id": "Liang-P-S", "name": { "family": "Liang", "given": "P. S." } }, { "id": "Wortman-Vaughan-J", "name": { "family": "Wortman Vaughan", "given": "J." }, "orcid": "0000-0002-7807-2018" } ] }, "primary_object": { "basename": "GAO_ANIPS_2021.pdf", "url": "https://authors.library.caltech.edu/records/jmn81-k7q70/files/GAO_ANIPS_2021.pdf" }, "resource_type": "book_section", "pub_year": "2021", "author_list": "Gao, Angela F.; Castellanos, Jorge C.; et el." }, { "id": "https://authors.library.caltech.edu/records/wbj75-1gv55", "eprint_id": 113765, "eprint_status": "archive", "datestamp": "2023-08-20 05:26:25", "lastmod": "2023-10-23 15:37:41", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Levis-Aviad", "name": { "family": "Levis", "given": "Aviad" }, "orcid": "0000-0001-7307-632X" }, { "id": "Lee-Daeyoung", "name": { "family": "Lee", "given": "Daeyoung" } }, { "id": "Tropp-J-A", "name": { "family": "Tropp", "given": "Joel A." }, "orcid": "0000-0003-1024-1791" }, { "id": "Gammie-Charles-F", "name": { "family": "Gammie", "given": "Charles F." }, "orcid": "0000-0001-7451-8935" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" } ] }, "title": "Inference of Black Hole Fluid-Dynamics from Sparse Interferometric Measurements", "ispublished": "unpub", "full_text_status": "restricted", "note": "\u00a9 2021 IEEE. \n\nThe authors would like to thank George Wong for his help with GRMHD simulations. AL is supported by the Zuckerman and Viterbi postdoctoral fellowships. This work was supported by NSF award 1935980: \"Next Generation Event Horizon Telescope Design,\" and Beyond Limits, and NSF awards 1743747, 1716327, and 2034306, XSEDE allocation TG-AST170024, and TACC Frontera LSCP AST20023. JAT was supported by ONR BRC Award N00014-18-1-2363 and NSF FRG Award 1952735.", "abstract": "We develop an approach to recover the underlying properties of fluid-dynamical processes from sparse measurements. We are motivated by the task of imaging the stochastically evolving environment surrounding black holes, and demonstrate how flow parameters can be estimated from sparse interferometric measurements used in radio astronomical imaging. To model the stochastic flow we use spatio-temporal Gaussian Random Fields (GRFs). The high dimensionality of the underlying source video makes direct representation via a GRF's full covariance matrix intractable. In contrast, stochastic partial differential equations are able to capture correlations at multiple scales by specifying only local interaction coefficients. Our approach estimates the coefficients of a space-time diffusion equation that dictates the stationary statistics of the dynamical process. We analyze our approach on realistic simulations of black hole evolution and demonstrate its advantage over state-of-the-art dynamic black hole imaging techniques.", "date": "2021-10", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "2320-2329", "id_number": "CaltechAUTHORS:20220307-188412000", "isbn": "978-1-6654-2812-5", "book_title": "2021 IEEE/CVF International Conference on Computer Vision (ICCV)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20220307-188412000", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "Zuckerman STEM Leadership Program" }, { "agency": "Viterbi fellowship" }, { "agency": "NSF", "grant_number": "AST-1935980" }, { "agency": "NSF", "grant_number": "OISE-1743747" }, { "agency": "NSF", "grant_number": "AST-1716327" }, { "agency": "NSF", "grant_number": "AST-2034306" }, { "agency": "NSF", "grant_number": "TG-AST170024" }, { "agency": "NSF", "grant_number": "AST-20023" }, { "agency": "Office of Naval Research (ONR)", "grant_number": "N00014-18-1-2363" }, { "agency": "NSF", "grant_number": "IIS-1952735" } ] }, "doi": "10.1109/iccv48922.2021.00234", "resource_type": "book_section", "pub_year": "2021", "author_list": "Levis, Aviad; Lee, Daeyoung; et el." }, { "id": "https://authors.library.caltech.edu/records/k65xp-y9147", "eprint_id": 109060, "eprint_status": "archive", "datestamp": "2023-08-20 05:13:01", "lastmod": "2023-10-23 17:32:16", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Cosner-Ryan-K", "name": { "family": "Cosner", "given": "Ryan K." } }, { "id": "Singletary-Andrew-W", "name": { "family": "Singletary", "given": "Andrew W." }, "orcid": "0000-0001-6635-4256" }, { "id": "Taylor-Andrew-J", "name": { "family": "Taylor", "given": "Andrew J." }, "orcid": "0000-0002-5990-590X" }, { "id": "Moln\u00e1r-Tam\u00e1s-G", "name": { "family": "Molnar", "given": "Tam\u00e1s G." }, "orcid": "0000-0002-9379-7121" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Ames-A-D", "name": { "family": "Ames", "given": "Aaron D." }, "orcid": "0000-0003-0848-3177" } ] }, "title": "Measurement-Robust Control Barrier Functions: Certainty in Safety with Uncertainty in State", "ispublished": "unpub", "full_text_status": "public", "note": "\u00a9 2021 IEEE. \n\nThis research is supported in part by the National Science Foundation, CPS Award #1932091; DOW Chemical, project 227027AT; British Petroleum; and Aerovironment.\n\nSubmitted - 2104.14030.pdf
", "abstract": "The increasing complexity of modern robotic systems and the environments they operate in necessitates the formal consideration of safety in the presence of imperfect measurements. In this paper we propose a rigorous framework for safety-critical control of systems with erroneous state estimates. We develop this framework by leveraging Control Barrier Functions (CBFs) and unifying the method of Backup Sets for synthesizing control invariant sets with robustness requirements\u2014the end result is the synthesis of Measurement-Robust Control Barrier Functions (MR-CBFs). This provides theoretical guarantees on safe behavior in the presence of imperfect measurements and improved robustness over standard CBF approaches. We demonstrate the efficacy of this framework both in simulation and experimentally on a Segway platform using an onboard stereo-vision camera for state estimation.", "date": "2021-09-27", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "6286-6291", "id_number": "CaltechAUTHORS:20210510-141401087", "isbn": "978-1-6654-1715-0", "book_title": "2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20210510-141401087", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "CNS-1932091" }, { "agency": "Dow Chemical Company", "grant_number": "227027AT" }, { "agency": "British Petroleum" }, { "agency": "AeroVironment" } ] }, "doi": "10.1109/IROS51168.2021.9636584", "primary_object": { "basename": "2104.14030.pdf", "url": "https://authors.library.caltech.edu/records/k65xp-y9147/files/2104.14030.pdf" }, "resource_type": "book_section", "pub_year": "2021", "author_list": "Cosner, Ryan K.; Singletary, Andrew W.; et el." }, { "id": "https://authors.library.caltech.edu/records/7249p-fqn69", "eprint_id": 109398, "eprint_status": "archive", "datestamp": "2023-08-20 03:17:38", "lastmod": "2024-01-15 21:22:55", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Sun-He", "name": { "family": "Sun", "given": "He" }, "orcid": "0000-0003-1526-6787" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" } ] }, "title": "Deep Probabilistic Imaging: Uncertainty Quantification and Multi-modal Solution Characterization for Computational Imaging", "ispublished": "unpub", "full_text_status": "public", "keywords": "Computational Photography, Image & Video Synthesis", "note": "\u00a9 2021 Association for the Advancement of Artificial Intelligence. \n\nPublished 2021-05-18. \n\nThis work was supported by NSF award 1935980: Next Generation Event Horizon Telescope Design, and Beyond Limits. The authors would also like to thank Joe Marino, Dominic Pesce, S. Kevin Zhou, and Tianwei Yin for the helpful discussions.\n\nAccepted Version - 2010.14462.pdf
Supplemental Material - DPIsupplement.pdf
", "abstract": "Computational image reconstruction algorithms generally produce a single image without any measure of uncertainty or confidence. Regularized Maximum Likelihood (RML) and feed-forward deep learning approaches for inverse problems typically focus on recovering a point estimate. This is a serious limitation when working with under-determined imaging systems, where it is conceivable that multiple image modes would be consistent with the measured data. Characterizing the space of probable images that explain the observational data is therefore crucial. In this paper, we propose a variational deep probabilistic imaging approach to quantify reconstruction uncertainty. Deep Probabilistic Imaging (DPI) employs an untrained deep generative model to estimate a posterior distribution of an unobserved image. This approach does not require any training data; instead, it optimizes the weights of a neural network to generate image samples that fit a particular measurement dataset. Once the network weights have been learned, the posterior distribution can be efficiently sampled. We demonstrate this approach in the context of interferometric radio imaging, which is used for black hole imaging with the Event Horizon Telescope, and compressed sensing Magnetic Resonance Imaging (MRI).", "date": "2021-05-18", "date_type": "published", "publisher": "Association for the Advancement of Artificial Intelligence", "id_number": "CaltechAUTHORS:20210604-142552450", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20210604-142552450", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "AST-1935980" }, { "agency": "Beyond Limits" } ] }, "local_group": { "items": [ { "id": "Astronomy-Department" } ] }, "doi": "10.48550/arXiv.2010.14462", "primary_object": { "basename": "2010.14462.pdf", "url": "https://authors.library.caltech.edu/records/7249p-fqn69/files/2010.14462.pdf" }, "related_objects": [ { "basename": "DPIsupplement.pdf", "url": "https://authors.library.caltech.edu/records/7249p-fqn69/files/DPIsupplement.pdf" } ], "resource_type": "book_section", "pub_year": "2021", "author_list": "Sun, He and Bouman, Katherine L." }, { "id": "https://authors.library.caltech.edu/records/r5pjt-ezf04", "eprint_id": 103735, "eprint_status": "archive", "datestamp": "2023-08-22 04:34:22", "lastmod": "2023-10-20 16:40:00", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Sun-He", "name": { "family": "Sun", "given": "He" } }, { "id": "Dalca-Adrian-V", "name": { "family": "Dalca", "given": "Adrian V." }, "orcid": "0000-0002-8422-0136" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" } ] }, "title": "Learning a Probabilistic Strategy for Computational Imaging Sensor Selection", "ispublished": "unpub", "full_text_status": "public", "keywords": "Computational Imaging, Optimized Sensing, Ising Model, Deep Learning, VLBI, Interferometry", "note": "\u00a9 2020 IEEE. \n\nThe authors would like to thank Lindy Blackburn, Alexander Raymond, Michael Johnson, and Sheperd Doeleman for helpful discussions on the constraints of a next-generation EHT array, and Michael Kellman for helpful discussions on Fourier ptychography. \n\nThis work was supported by NSF award 1935980: \"Next Generation Event Horizon Telescope Design,\" and Beyond Limits.\n\nAccepted Version - 2003.10424.pdf
", "abstract": "Optimized sensing is important for computational imaging in low-resource environments, when images must be recovered from severely limited measurements. In this paper, we propose a physics-constrained, fully differentiable, autoencoder that learns a probabilistic sensor-sampling strategy for optimized sensor design. The proposed method learns a system's preferred sampling distribution that characterizes the correlations between different sensor selections as a binary, fully-connected Ising model. The learned probabilistic model is achieved by using a Gibbs sampling inspired network architecture, and is trained end-to-end with a reconstruction network for efficient co-design. The proposed framework is applicable to sensor selection problems in a variety of computational imaging applications. In this paper, we demonstrate the approach in the context of a very-long-baseline-interferometry (VLBI) array design task, where sensor correlations and atmospheric noise present unique challenges. We demonstrate results broadly consistent with expectation, and draw attention to particular structures preferred in the telescope array geometry that can be leveraged to plan future observations and design array expansions.", "date": "2020-04", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "1-12", "id_number": "CaltechAUTHORS:20200605-134947553", "isbn": "9781728152301", "book_title": "2020 IEEE International Conference on Computational Photography (ICCP)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20200605-134947553", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "AST-1935980" } ] }, "local_group": { "items": [ { "id": "Astronomy-Department" } ] }, "doi": "10.1109/iccp48838.2020.9105133", "primary_object": { "basename": "2003.10424.pdf", "url": "https://authors.library.caltech.edu/records/r5pjt-ezf04/files/2003.10424.pdf" }, "resource_type": "book_section", "pub_year": "2020", "author_list": "Sun, He; Dalca, Adrian V.; et el." }, { "id": "https://authors.library.caltech.edu/records/5gqc8-zfw53", "eprint_id": 94482, "eprint_status": "archive", "datestamp": "2023-08-22 01:41:13", "lastmod": "2023-10-20 18:00:01", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Ye-Vickie", "name": { "family": "Ye", "given": "Vickie" } }, { "id": "Yedidia-A-B", "name": { "family": "Yedidia", "given": "Adam B." } }, { "id": "Durand-F", "name": { "family": "Durand", "given": "Fredo" } }, { "id": "Wornell-G-W", "name": { "family": "Wornell", "given": "Gregory W." } }, { "id": "Torralba-A", "name": { "family": "Torralba", "given": "Antonio" } }, { "id": "Freeman-W-T", "name": { "family": "Freeman", "given": "William T." }, "orcid": "0000-0002-2231-7995" } ] }, "title": "Turning Corners into Cameras: Principles and Methods", "ispublished": "unpub", "full_text_status": "restricted", "note": "\u00a9 2017 IEEE. \n\nThis work was supported in part by the DARPA REVEAL Program under contract No. HR0011-16-C-0030, NSF Grant 1212849, Shell Research, and an NDSEG Fellowship (to ABY). We thank Yoav Schechner, Jeff Shapiro, Franco Wang, and Vivek Goyal for helpful discussions.", "abstract": "We show that walls, and other obstructions with edges, can be exploited as naturally-occurring \"cameras\" that reveal the hidden scenes beyond them. In particular, we demonstrate methods for using the subtle spatio-temporal radiance variations that arise on the ground at the base of a wall's edge to construct a one-dimensional video of the hidden scene behind the wall. The resulting technique can be used for a variety of applications in diverse physical settings. From standard RGB video recordings, we use edge cameras to recover 1-D videos that reveal the number and trajectories of people moving in an occluded scene. We further show that adjacent wall edges, such as those that arise in the case of an open doorway, yield a stereo camera from which the 2-D location of hidden, moving objects can be recovered. We demonstrate our technique in a number of indoor and outdoor environments involving varied floor surfaces and illumination conditions.", "date": "2017-10", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "2289-2297", "id_number": "CaltechAUTHORS:20190404-161219288", "isbn": "9781538610329", "book_title": "2017 IEEE International Conference on Computer Vision (ICCV)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190404-161219288", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "Defense Advanced Research Projects Agency (DARPA)", "grant_number": "HR0011-16-C-0030" }, { "agency": "NSF", "grant_number": "IIS-1212849" }, { "agency": "Shell Research" }, { "agency": "National Defense Science and Engineering Graduate (NDSEG) Fellowship" } ] }, "local_group": { "items": [ { "id": "Astronomy-Department" } ] }, "doi": "10.1109/iccv.2017.249", "resource_type": "book_section", "pub_year": "2017", "author_list": "Bouman, Katherine L.; Ye, Vickie; et el." }, { "id": "https://authors.library.caltech.edu/records/mrsap-xv786", "eprint_id": 94483, "eprint_status": "archive", "datestamp": "2023-08-22 01:35:23", "lastmod": "2023-10-20 18:00:04", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Sreehari-S", "name": { "family": "Sreehari", "given": "Suhas" } }, { "id": "Venkatakrishnan-S-V", "name": { "family": "Venkatakrishnan", "given": "S. V." } }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Simmons-J-P", "name": { "family": "Simmons", "given": "Jeff P." } }, { "id": "Drummy-L-F", "name": { "family": "Drummy", "given": "Larry F." }, "orcid": "0000-0002-6452-5768" }, { "id": "Bouman-C-A", "name": { "family": "Bouman", "given": "Charles A." } } ] }, "title": "Multi-Resolution Data Fusion for Super-Resolution Electron Microscopy", "ispublished": "unpub", "full_text_status": "restricted", "note": "\u00a9 2017 IEEE.", "abstract": "Perhaps surprisingly, all electron microscopy (EM) data collected to date is less than a cubic millimeter - presenting a huge demand in the materials and biological sciences to image at greater speed and lower dosage, while maintaining resolution. Traditional EM imaging based on homogeneous raster scanning severely limits the volume of high-resolution data that can be collected, and presents a fundamental limitation to understanding physical processes such as material deformation and crack propagation. We introduce a multi-resolution data fusion (MDF) method for super-resolution computational EM. Our method combines innovative data acquisition with novel algorithmic techniques to dramatically improve the resolution/volume/speed trade-off. The key to our approach is to collect the entire sample at low resolution, while simultaneously collecting a small fraction of data at high resolution. The high-resolution measurements are then used to create a material-specific model that is used within the \"plug-and-play\" framework to dramatically improve resolution of the low-resolution data. We present results using FEI electron microscope data that demonstrate super-resolution factors of 4x-16x, while substantially maintaining high image quality and reducing dosage.", "date": "2017-07", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "1084-1092", "id_number": "CaltechAUTHORS:20190404-161219383", "isbn": "9781538607336", "book_title": "2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190404-161219383", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "local_group": { "items": [ { "id": "Astronomy-Department" } ] }, "doi": "10.1109/cvprw.2017.146", "resource_type": "book_section", "pub_year": "2017", "author_list": "Sreehari, Suhas; Venkatakrishnan, S. V.; et el." }, { "id": "https://authors.library.caltech.edu/records/08h0f-azh33", "eprint_id": 94521, "eprint_status": "archive", "datestamp": "2023-08-22 01:31:29", "lastmod": "2024-01-14 21:42:05", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Dalca-A-V", "name": { "family": "Dalca", "given": "Adrian V." }, "orcid": "0000-0002-8422-0136" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Freeman-W-T", "name": { "family": "Freeman", "given": "William T." }, "orcid": "0000-0002-2231-7995" }, { "id": "Rost-N-S", "name": { "family": "Rost", "given": "Natalia S." } }, { "id": "Sabuncu-M-R", "name": { "family": "Sabuncu", "given": "Mert R." }, "orcid": "0000-0002-7068-719X" }, { "id": "Golland-P", "name": { "family": "Golland", "given": "Polina" } } ] }, "title": "Population Based Image Imputation", "ispublished": "unpub", "full_text_status": "restricted", "note": "\u00a9 Springer International Publishing AG 2017. \n\nWe acknowledge the following funding sources: NIH NINDS R01NS086905, NIH NICHD U01HD087211, NIH NIBIB NAC P41EB015902, NIH R41AG052246-01, 1K25EB013649-01, 1R21AG050122-01, and Wistron Corporation.", "abstract": "We present an algorithm for creating high resolution anatomically plausible images consistent with acquired clinical brain MRI scans with large inter-slice spacing. Although large databases of clinical images contain a wealth of information, medical acquisition constraints result in sparse scans that miss much of the anatomy. These characteristics often render computational analysis impractical as standard processing algorithms tend to fail when applied to such images. Highly specialized or application-specific algorithms that explicitly handle sparse slice spacing do not generalize well across problem domains. In contrast, our goal is to enable application of existing algorithms that were originally developed for high resolution research scans to significantly undersampled scans. We introduce a model that captures fine-scale anatomical similarity across subjects in clinical image collections and use it to fill in the missing data in scans with large slice spacing. Our experimental results demonstrate that the proposed method outperforms current upsampling methods and promises to facilitate subsequent analysis not previously possible with scans of this quality.", "date": "2017-05-23", "date_type": "published", "publisher": "Springer", "place_of_pub": "Cham, Switzerland", "pagerange": "659-671", "id_number": "CaltechAUTHORS:20190405-160617876", "isbn": "9783319590493", "book_title": "Information Processing in Medical Imaging. IPMI 2017", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190405-160617876", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NIH", "grant_number": "R01NS086905" }, { "agency": "NIH", "grant_number": "U01HD087211" }, { "agency": "NIH", "grant_number": "P41EB015902" }, { "agency": "NIH", "grant_number": "R41AG052246-01" }, { "agency": "NIH", "grant_number": "1K25EB013649-01" }, { "agency": "NIH", "grant_number": "1R21AG050122-01" }, { "agency": "Wistron Corporation" } ] }, "local_group": { "items": [ { "id": "Astronomy-Department" } ] }, "contributors": { "items": [ { "id": "Niethammer-M", "name": { "family": "Niethammer", "given": "Marc" } }, { "id": "Styner-M", "name": { "family": "Styner", "given": "Martin" } }, { "id": "Aylward-S", "name": { "family": "Aylward", "given": "Stephen" } }, { "id": "Zhu-Hongtu", "name": { "family": "Zhu", "given": "Hongtu" } }, { "id": "Oguz-I", "name": { "family": "Oguz", "given": "Ipek" } }, { "id": "Yap-Pew-Thian", "name": { "family": "Yap", "given": "Pew-Thian" } }, { "id": "Shen-Dinggang", "name": { "family": "Shen", "given": "Dinggang" } } ] }, "doi": "10.1007/978-3-319-59050-9_52", "resource_type": "book_section", "pub_year": "2017", "author_list": "Dalca, Adrian V.; Bouman, Katherine L.; et el." }, { "id": "https://authors.library.caltech.edu/records/h9qvb-h4m17", "eprint_id": 94522, "eprint_status": "archive", "datestamp": "2023-08-19 00:22:40", "lastmod": "2023-10-20 18:02:14", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Xue-Tianfan", "name": { "family": "Xue", "given": "Tianfan" } }, { "id": "Wu-Jiajun", "name": { "family": "Wu", "given": "Jiajun" }, "orcid": "0000-0002-4176-343X" }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Freeman-W-T", "name": { "family": "Freeman", "given": "William T." }, "orcid": "0000-0002-2231-7995" } ] }, "title": "Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks", "ispublished": "unpub", "full_text_status": "public", "note": "\u00a9 2016 Neural Information Processing Systems Foundation. \n\nThe authors thank Yining Wang for helpful discussions. This work is supported by NSF Robust Intelligence 1212849, NSF Big Data 1447476, ONR MURI 6923196, Adobe, and Shell Research. The authors would also like to thank Nvidia for GPU donation. \n\nThe first two authors contributed equally to this work.\n\nPublished - 6552-visual-dynamics-probabilistic-future-frame-synthesis-via-cross-convolutional-networks.pdf
Supplemental Material - 6552-visual-dynamics-probabilistic-future-frame-synthesis-via-cross-convolutional-networks-supplemental.zip
", "abstract": "We study the problem of synthesizing a number of likely future frames from a single input image. In contrast to traditional methods, which have tackled this problem in a deterministic or non-parametric way, we propose a novel approach which models future frames in a probabilistic manner. Our proposed method is therefore able to synthesize multiple possible next frames using the same model. Solving this challenging problem involves low- and high-level image and motion understanding for successful image synthesis. Here, we propose a novel network structure, namely a Cross Convolutional Network, that encodes images as feature maps and motion information as convolutional kernels to aid in synthesizing future frames. In experiments, our model performs well on both synthetic data, such as 2D shapes and animated game sprites, as well as on real-wold video data. We show that our model can also be applied to tasks such as visual analogy-making, and present analysis of the learned network representations.", "date": "2016-12", "date_type": "published", "publisher": "Neural Information Processing Systems Foundation", "place_of_pub": "La Jolla, CA", "id_number": "CaltechAUTHORS:20190405-161634029", "isbn": "9781510838819", "book_title": "Advances in Neural Information Processing Systems (NIPS 2016)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190405-161634029", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "IIS-1212849" }, { "agency": "Office of Naval Research", "grant_number": "6923196" }, { "agency": "Adobe" }, { "agency": "Shell Research" }, { "agency": "nVidia" } ] }, "contributors": { "items": [ { "id": "Lee-Daniel-D", "name": { "family": "Lee", "given": "Daniel D." } }, { "id": "Sugiyama-Masashi", "name": { "family": "Sugiyama", "given": "Masashi" } }, { "id": "von-Luxburg-U", "name": { "family": "von Luxburg", "given": "Ulrike" } }, { "id": "Guyon-I", "name": { "family": "Guyon", "given": "Isabelle" } }, { "id": "Garnett-R", "name": { "family": "Garnett", "given": "Roman" } } ] }, "doi": "10.48550/arXiv.1607.02586", "primary_object": { "basename": "6552-visual-dynamics-probabilistic-future-frame-synthesis-via-cross-convolutional-networks-supplemental.zip", "url": "https://authors.library.caltech.edu/records/h9qvb-h4m17/files/6552-visual-dynamics-probabilistic-future-frame-synthesis-via-cross-convolutional-networks-supplemental.zip" }, "related_objects": [ { "basename": "6552-visual-dynamics-probabilistic-future-frame-synthesis-via-cross-convolutional-networks.pdf", "url": "https://authors.library.caltech.edu/records/h9qvb-h4m17/files/6552-visual-dynamics-probabilistic-future-frame-synthesis-via-cross-convolutional-networks.pdf" } ], "resource_type": "book_section", "pub_year": "2016", "author_list": "Xue, Tianfan; Wu, Jiajun; et el." }, { "id": "https://authors.library.caltech.edu/records/9at3s-nj868", "eprint_id": 94484, "eprint_status": "archive", "datestamp": "2023-08-20 11:59:12", "lastmod": "2023-10-20 18:00:09", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Johnson-M-D", "name": { "family": "Johnson", "given": "Michael D." }, "orcid": "0000-0002-4120-3029" }, { "id": "Zoran-D", "name": { "family": "Zoran", "given": "Daniel" } }, { "id": "Fish-V-L", "name": { "family": "Fish", "given": "Vincent L." }, "orcid": "0000-0002-7128-9345" }, { "id": "Doeleman-S-S", "name": { "family": "Doeleman", "given": "Sheperd S." }, "orcid": "0000-0002-9031-0904" }, { "id": "Freeman-W-T", "name": { "family": "Freeman", "given": "William T." }, "orcid": "0000-0002-2231-7995" } ] }, "title": "Computational Imaging for VLBI Image Reconstruction", "ispublished": "unpub", "full_text_status": "public", "note": "\u00a9 2016 IEEE. \n\nWe would like to thank Andrew Chael, Katherine Rosenfeld, and Lindy Blackburn for all of their helpful discussions and feedback. This work was partially supported by NSF CGV-1111415. Katherine Bouman was partially supported by an NSF Graduate Fellowship. We also thank the National Science Foundation (AST-1310896, AST-1211539, and AST-1440254) and the Gordon and Betty Moore Foundation (GBMF-3561) for financial support of this work. This study makes use of 43 GHz VLBA data from the VLBA-BU Blazar Monitoring Program (VLBA-BU-BLAZAR; http://www.bu.edu/blazars/VLBAproject.html), funded by NASA through the Fermi Guest Investigator Program. The VLBA is an instrument of the National Radio Astronomy Observatory. The National Radio Astronomy Observatory is a facility of the National Science Foundation operated by Associated Universities, Inc.\n\nAccepted Version - 1512.01413.pdf
", "abstract": "Very long baseline interferometry (VLBI) is a technique for imaging celestial radio emissions by simultaneously observing a source from telescopes distributed across Earth. The challenges in reconstructing images from fine angular resolution VLBI data are immense. The data is extremely sparse and noisy, thus requiring statistical image models such as those designed in the computer vision community. In this paper we present a novel Bayesian approach for VLBI image reconstruction. While other methods often require careful tuning and parameter selection for different types of data, our method (CHIRP) produces good results under different settings such as low SNR or extended emission. The success of our method is demonstrated on realistic synthetic experiments as well as publicly available real data. We present this problem in a way that is accessible to members of the community, and provide a dataset website (vlbiimaging.csail.mit.edu) that facilitates controlled comparisons across algorithms.", "date": "2016-06", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "913-922", "id_number": "CaltechAUTHORS:20190404-161219475", "isbn": "9781467388511", "book_title": "2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190404-161219475", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "CGV-1111415" }, { "agency": "NSF Graduate Research Fellowship" }, { "agency": "NSF", "grant_number": "AST-1310896" }, { "agency": "NSF", "grant_number": "AST-1211539" }, { "agency": "NSF", "grant_number": "AST-1440254" }, { "agency": "Gordon and Betty Moore Foundation", "grant_number": "GBMF-3561" }, { "agency": "NASA" } ] }, "doi": "10.1109/cvpr.2016.105", "primary_object": { "basename": "1512.01413.pdf", "url": "https://authors.library.caltech.edu/records/9at3s-nj868/files/1512.01413.pdf" }, "resource_type": "book_section", "pub_year": "2016", "author_list": "Bouman, Katherine L.; Johnson, Michael D.; et el." }, { "id": "https://authors.library.caltech.edu/records/k2yby-6dy13", "eprint_id": 94523, "eprint_status": "archive", "datestamp": "2023-08-19 22:27:48", "lastmod": "2023-10-20 18:02:17", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Xiao-Bei", "name": { "family": "Xiao", "given": "Bei" } }, { "id": "Battaglia-P", "name": { "family": "Battaglia", "given": "Peter" } }, { "id": "Freeman-W-T", "name": { "family": "Freeman", "given": "William T." }, "orcid": "0000-0002-2231-7995" } ] }, "title": "Estimating the Material Properties of Fabric from Video", "ispublished": "unpub", "full_text_status": "public", "note": "\u00a9 2013 IEEE. \n\nWe would like to thank Lowell ACMTL, particularly Patrick Drane, for their help in collecting data for this project. We would also like to thank Adrian Dalca for all of his helpful discussions and feedback. This work was partially supported by NSF CGV-1111415 and NSF CGV-1212928. Katherine Bouman was partially supported by an NSF Graduate Fellow-ship. Bei Xiao was supported by an MIT Intelligent Initiative Postdoctoral Fellowship. Peter Battaglia was supported by (IARPA) - D10PC20023.\n\nAccepted Version - Bouman_Estimating_the_Material_2013_ICCV_paper.pdf
", "abstract": "Passively estimating the intrinsic material properties of deformable objects moving in a natural environment is essential for scene understanding. We present a framework to automatically analyze videos of fabrics moving under various unknown wind forces, and recover two key material properties of the fabric: stiffness and area weight. We extend features previously developed to compactly represent static image textures to describe video textures, such as fabric motion. A discriminatively trained regression model is then used to predict the physical properties of fabric from these features. The success of our model is demonstrated on a new, publicly available database of fabric videos with corresponding measured ground truth material properties. We show that our predictions are well correlated with ground truth measurements of stiffness and density for the fabrics. Our contributions include: (a) a database that can be used for training and testing algorithms for passively predicting fabric properties from video, (b) an algorithm for predicting the material properties of fabric from a video, and (c) a perceptual study of humans' ability to estimate the material properties of fabric from videos and images.", "date": "2013-12", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "1984-1991", "id_number": "CaltechAUTHORS:20190405-163101345", "isbn": "9781479928408", "book_title": "2013 IEEE International Conference on Computer Vision", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190405-163101345", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "NSF", "grant_number": "CGV-1111415" }, { "agency": "NSF", "grant_number": "CGV-1212928" }, { "agency": "NSF Graduate Research Fellowship" }, { "agency": "Massachusetts Institute of Technology (MIT)" }, { "agency": "Intelligence Advanced Research Projects Activity (IARPA)", "grant_number": "D10PC20023" } ] }, "doi": "10.1109/iccv.2013.455", "primary_object": { "basename": "Bouman_Estimating_the_Material_2013_ICCV_paper.pdf", "url": "https://authors.library.caltech.edu/records/k2yby-6dy13/files/Bouman_Estimating_the_Material_2013_ICCV_paper.pdf" }, "resource_type": "book_section", "pub_year": "2013", "author_list": "Bouman, Katherine L.; Xiao, Bei; et el." }, { "id": "https://authors.library.caltech.edu/records/88fbc-3ex36", "eprint_id": 94485, "eprint_status": "archive", "datestamp": "2023-08-19 13:23:41", "lastmod": "2023-10-20 18:00:18", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Ni-Karl", "name": { "family": "Ni", "given": "Karl" } }, { "id": "Phelps-E", "name": { "family": "Phelps", "given": "Ethan" } }, { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Bliss-N", "name": { "family": "Bliss", "given": "Nadya" } } ] }, "title": "Training image classifiers with similarity metrics, linear programming, and minimal supervision", "ispublished": "unpub", "full_text_status": "restricted", "note": "\u00a9 2012 IEEE. \n\nWe would like to thank Andrew Bolstad at MIT Lincoln Laboratory for all the help, advice, and good ideas he has given us in regard to convex optimization for sparse regularization techniques. This work is sponsored by the Assistant Secretary of Defense for Research & Engineering under Air Force Contract # FA8721-05-C-0002. Opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the United States Government. \n\nWe would like to thank Andrew Bolstad at MIT Lincoln Laboratory for all the help, advice, and good ideas he has given us in regard to convex optimization for sparse regularization techniques.", "abstract": "Image classification is a classical computer vision problem with applications to semantic image annotation, querying, and indexing. Recent and effective generative techniques assume Gaussianity, rely on distance metrics, and estimate distributions, but are unfortunately not convex nor keep computational architecture in mind. We propose image content classification through convex linear programming using similarity metrics rather than commonly-used Mahalanobis distances. The algorithm is solved through a hybrid iterative method that takes advantage of optimization space properties. Our optimization problem uses dot products in the feature space exclusively, and therefore can be extended to non-linear kernel functions in the transductive setting.", "date": "2012-11", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "1979-1983", "id_number": "CaltechAUTHORS:20190404-161219565", "isbn": "9781467350518", "book_title": "2012 Conference Record of the Forty Sixth Asilomar Conference on Signals, Systems and Computers (ASILOMAR)", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190404-161219565", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "Air Force Office of Scientific Research (AFOSR)", "grant_number": "FA8721-05-C-0002" } ] }, "doi": "10.1109/acssc.2012.6489386", "resource_type": "book_section", "pub_year": "2012", "author_list": "Ni, Karl; Phelps, Ethan; et el." }, { "id": "https://authors.library.caltech.edu/records/br7bc-cr246", "eprint_id": 94526, "eprint_status": "archive", "datestamp": "2023-08-19 05:19:30", "lastmod": "2024-01-14 21:42:07", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine" }, "orcid": "0000-0003-0077-4367" }, { "id": "Ramachandra-V", "name": { "family": "Ramachandra", "given": "Vikas" } }, { "id": "Atanassov-K", "name": { "family": "Atanassov", "given": "Kalin" } }, { "id": "Aleksic-M", "name": { "family": "Aleksic", "given": "Mickey" } }, { "id": "Goma-S-R", "name": { "family": "Goma", "given": "Sergio R." } } ] }, "title": "RAW camera DPCM compression performance analysis", "ispublished": "unpub", "full_text_status": "public", "keywords": "MIPI, DPCM, RAW, compression, MTF", "note": "\u00a9 2011 Society of Photo-Optical Instrumentation Engineers (SPIE).\n\nPublished - 78670N.pdf
", "abstract": "The MIPI standard has adopted DPCM compression for RAW data images streamed from mobile cameras. This DPCM is line based and uses either a simple 1 or 2 pixel predictor. In this paper, we analyze the DPCM compression performance as MTF degradation. To test this scheme's performance, we generated Siemens star images and binarized them to 2-level images. These two intensity values where chosen such that their intensity difference corresponds to those pixel differences which result in largest relative errors in the DPCM compressor. (E.g. a pixel transition from 0 to 4095 corresponds to an error of 6 between the DPCM compressed value and the original pixel value). The DPCM scheme introduces different amounts of error based on the pixel difference. We passed these modified Siemens star chart images to this compressor and compared the compressed images with the original images using IT3 MTF response plots for slanted edges. Further, we discuss the PSF influence on DPCM error and its propagation through the image processing pipe.", "date": "2011-01-24", "date_type": "published", "publisher": "Society of Photo-optical Instrumentation Engineers (SPIE)", "place_of_pub": "Bellingham, WA", "pagerange": "Art. No. 78670N", "id_number": "CaltechAUTHORS:20190405-164254906", "isbn": "9780819484048", "book_title": "Image Quality and System Performance VIII", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190405-164254906", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "contributors": { "items": [ { "id": "Farnand-S-P", "name": { "family": "Farnand", "given": "Susan P." } }, { "id": "Gaykema-F", "name": { "family": "Gaykema", "given": "Frans" } } ] }, "doi": "10.1117/12.872637", "primary_object": { "basename": "78670N.pdf", "url": "https://authors.library.caltech.edu/records/br7bc-cr246/files/78670N.pdf" }, "resource_type": "book_section", "pub_year": "2011", "author_list": "Bouman, Katherine; Ramachandra, Vikas; et el." }, { "id": "https://authors.library.caltech.edu/records/qkbw8-03e95", "eprint_id": 94487, "eprint_status": "archive", "datestamp": "2023-08-19 01:53:14", "lastmod": "2023-10-20 18:00:26", "type": "book_section", "metadata_visibility": "show", "creators": { "items": [ { "id": "Bouman-K-L", "name": { "family": "Bouman", "given": "Katherine L." }, "orcid": "0000-0003-0077-4367" }, { "id": "Abdollahian-G", "name": { "family": "Abdollahian", "given": "Golnaz" } }, { "id": "Boutin-M", "name": { "family": "Boutin", "given": "Mireille" } }, { "id": "Delp-E-J", "name": { "family": "Delp", "given": "Edward J." } } ] }, "title": "A low complexity method for detection of text area in natural images", "ispublished": "unpub", "full_text_status": "public", "keywords": "Text Detection, Text Segmentation, Sign Detection, Mobile Devices", "note": "\u00a9 2010 IEEE. \n\nThis work was sponsored by Next Wave Systems, LLC.\n\nPublished - 05495331.pdf
", "abstract": "We propose a low complexity method for segmentation of text regions in natural images. This algorithm is designed for mobile applications (e.g. unmanned or hand-held devices) in which computational and energy resources are limited. No prior assumption is made regarding the text size, font, language, character set or the camera angle. However, the text is assumed to be located on a piecewise homogeneous background with a contrasting color. We have deployed our method on a Nokia N800 Internet tablet as part of a system for automatic detection and translation of outdoor signs. Our experiments show that the 0.3 megapixel images taken by the phone camera can be accurately segmented within the device in a fraction of a second.", "date": "2010-03", "date_type": "published", "publisher": "IEEE", "place_of_pub": "Piscataway, NJ", "pagerange": "1050-1053", "id_number": "CaltechAUTHORS:20190404-161219749", "isbn": "9781424442959", "book_title": "2010 IEEE International Conference on Acoustics, Speech and Signal Processing", "official_url": "https://resolver.caltech.edu/CaltechAUTHORS:20190404-161219749", "rights": "No commercial reproduction, distribution, display or performance rights in this work are provided.", "funders": { "items": [ { "agency": "Next Wave Systems" } ] }, "doi": "10.1109/icassp.2010.5495331", "primary_object": { "basename": "05495331.pdf", "url": "https://authors.library.caltech.edu/records/qkbw8-03e95/files/05495331.pdf" }, "resource_type": "book_section", "pub_year": "2010", "author_list": "Bouman, Katherine L.; Abdollahian, Golnaz; et el." } ]