Jason J. Corso
| Snippets by Topic | |
| * | Active Clustering |
| * | Activity Recognition |
| * | Medical Imaging |
| * | Metric Learning |
| * | Semantic Segmentation |
| * | Video Segmentation |
| * | Video Understanding |
| Selected Project Pages | |
| * | Action Bank |
| * | LIBSVX: Supervoxel Library and Evaluation |
| * | Brain Tumor Segmentation |
| * | CAREER: Generalized Image Understanding |
| * | Summer of Code 2010: The Visual Noun |
| * | ACE: Active Clustering |
| * | ISTARE: Intelligent Spatiotemporal Activity Reasoning Engine |
| * | GBS: Guidance by Semantics |
| * | Semantic Video Summarization |
| Data Sets | |
| * | YouCook |
| * | Chen Xiph.org |
| * | UB/College Park Building Facades |
| Other Information | |
| * | Code/Data Downloads |
| * | List of Grants |
Semantic Region Labeling
People: Jason Corso, Albert Chen
Past Collaborators: Zhuowen Tu (UCLA), Alan Yuille (UCLA) Natural scene understanding, segmenting and labeling image regions with semantically meaningful labels (e.g., trees, cars, etc.), has increasingly attracted attention since it is a key aspect in image understanding and image search. In general, region labeling is a very hard problem due to the large variation of natural images ranging from indoor to outdoor, small to large scale, and rural to city scenes. Moreover, some type of objects, e.g. buildings, may have very different designs and appear very differently under different viewing directions and scales. We have proposed a supervised approach for combined image segmentation and region labeling. We use a hybrid discriminative-generative modeling scheme. The discriminative term is modeled by an extension of the probabilistic boosting tree algorithm that does multi-class representation in a tree structure. It can better handle the variability in natural image patches. The generative terms captures the local context of the image regions. The method has been applied to medical images (see the brain tumor page for examples). It has been applied to natural photos. Here is a montage on the MSRC v2 dataset.
We've also worked with fused overhead imagery and LIDAR data for detecting buildings.
We're currently working on extensions of our methods to video.
Past Collaborators: Zhuowen Tu (UCLA), Alan Yuille (UCLA) Natural scene understanding, segmenting and labeling image regions with semantically meaningful labels (e.g., trees, cars, etc.), has increasingly attracted attention since it is a key aspect in image understanding and image search. In general, region labeling is a very hard problem due to the large variation of natural images ranging from indoor to outdoor, small to large scale, and rural to city scenes. Moreover, some type of objects, e.g. buildings, may have very different designs and appear very differently under different viewing directions and scales. We have proposed a supervised approach for combined image segmentation and region labeling. We use a hybrid discriminative-generative modeling scheme. The discriminative term is modeled by an extension of the probabilistic boosting tree algorithm that does multi-class representation in a tree structure. It can better handle the variability in natural image patches. The generative terms captures the local context of the image regions. The method has been applied to medical images (see the brain tumor page for examples). It has been applied to natural photos. Here is a montage on the MSRC v2 dataset.
We've also worked with fused overhead imagery and LIDAR data for detecting buildings.
We're currently working on extensions of our methods to video.
Publications
| [1] | P. Das, R. K. Srihari, and J. J. Corso. Translating related words to videos and back through latent topics. In Proceedings of Sixth ACM International Conference on Web Search and Data Mining, 2013. [ bib | .pdf ] |
| [2] | Y. Miao and J. J. Corso. Hamiltonian streamline guided feature extraction with application to face detection. Journal of Neurocomputing, 120:226-234, 2013. Early version appears as arXiv.org tech report 1108.3525v1. [ bib | http ] |
| [3] | J. J. Corso. Toward parts-based scene understanding with pixel-support parts-sparse pictorial structures. Pattern Recognition Letters: Special Issue on Scene Understanding and Behavior Analysis, 34(7):762-769, 2013. Early version appears as arXiv.org tech report 1108.4079v1. [ bib | .pdf ] |
| [4] | J. Ryde and J. J. Corso. Fast voxel maps with counting bloom filters. In Proceedings of International Conference on Intelligent Robots and Systems, 2012. [ bib | code | .pdf ] |
| [5] | A. Y. C. Chen and J. J. Corso. Temporally consistent multi-class video-object segmentation with the video graph-shifts algorithm. In Proceedings of the 2011 IEEE Workshop on Motion and Video Computing, 2011. [ bib | code | project | .pdf ] |
| [6] | A. Y. C. Chen and J. J. Corso. Propagating multi-class pixel labels throughout video frames. In Proceedings of Western New York Image Processing Workshop, 2010. [ bib | .pdf ] |
| [7] | A. Y. C. Chen, J. J. Corso, and L. Wang. HOPS: Efficient region labeling using higher order proxy neighborhoods. In Proceedings of International Conference on Pattern Recognition, 2008. [ bib | .pdf ] |
| [8] | J. J. Corso, A. Yuille, and Z. Tu. Graph-Shifts: Natural Image Labeling by Dynamic Hierarchical Computing. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2008. [ bib | code | project | .pdf ] |
| [9] | J. J. Corso. Discriminative Modeling by Boosting on Multilevel Aggregates. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2008. [ bib | .pdf ] |
| [10] | I. Nwogu and J. J. Corso. (BP)2: Beyond Pairwise Belief Propagation, Labeling by Approximating Kikuchi Free Energies. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2008. [ bib | .pdf ] |