Search For:

Displaying 1-16 out of 16 total
Discovering localized attributes for fine-grained recognition
Found in: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
By Kun Duan,D. Parikh,D. Crandall,K. Grauman
Issue Date:June 2012
pp. 3474-3481
Attributes are visual concepts that can be detected by machines, understood by humans, and shared across categories. They are particularly useful for fine-grained domains where categories are closely related to one other (e.g. bird species recognition). In...
 
Observe locally, infer globally: A space-time MRF for detecting abnormal activities with incremental updates
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Jaechul Kim, K. Grauman
Issue Date:June 2009
pp. 2921-2928
We propose a space-time Markov random field (MRF) model to detect abnormal activities in video. The nodes in the MRF graph correspond to a grid of local regions in the video frames, and neighboring nodes in both space and time are associated with links. To...
 
Learning the easy things first: Self-paced visual category discovery
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Yong Jae Lee,K. Grauman
Issue Date:June 2011
pp. 1721-1728
Objects vary in their visual complexity, yet existing discovery methods perform
 
Sharing features between objects and their attributes
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Sung Ju Hwang, Fei Sha,K. Grauman
Issue Date:June 2011
pp. 1761-1768
Visual attributes expose human-defined semantics to object recognition models, but existing work largely restricts their influence to mid-level cues during classifier training. Rather than treat attributes as intermediate features, we consider how learning...
 
Reading between the Lines: Object Localization Using Implicit Cues from Image Tags
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Sung Ju Hwang,K. Grauman
Issue Date:June 2012
pp. 1145-1158
Current uses of tagged images typically exploit only the most explicit information: the link between the nouns named and the objects present somewhere in the image. We propose to leverage “unspoken” cues that rest within an ordered list of image tags so as...
 
Geodesic flow kernel for unsupervised domain adaptation
Found in: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
By Boqing Gong, Yuan Shi, Fei Sha,K. Grauman
Issue Date:June 2012
pp. 2066-2073
In real-world applications of visual recognition, many factors - such as pose, illumination, or image quality - can cause a significant mismatch between the source domain on which classifiers are trained and the target domain to which those classifiers are...
 
WhittleSearch: Image search with relative attribute feedback
Found in: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
By A. Kovashka,D. Parikh,K. Grauman
Issue Date:June 2012
pp. 2973-2980
We propose a novel mode of feedback for image search, where a user describes which properties of exemplar images should be adjusted in order to more closely match his/her mental model of the image(s) sought. For example, perusing image results for a query ...
 
Efficient activity detection with max-subgraph search
Found in: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
By Chao-Yeh Chen,K. Grauman
Issue Date:June 2012
pp. 1274-1281
We propose an efficient approach that unifies activity categorization with space-time localization. The main idea is to pose activity detection as a maximum-weight connected subgraph problem over a learned space-time graph constructed on the test sequence....
 
Efficient region search for object detection
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By S. Vijayanarasimhan,K. Grauman
Issue Date:June 2011
pp. 1401-1408
We propose a branch-and-cut strategy for efficient region-based object detection. Given an oversegmented image, our method determines the subset of spatially contiguous regions whose collective features will maximize a classifier's score. We formulate the ...
 
Large-scale live active learning: Training object detectors with crawled data and crowds
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By S. Vijayanarasimhan,K. Grauman
Issue Date:June 2011
pp. 1449-1456
Active learning and crowdsourcing are promising ways to efficiently build up training sets for object recognition, but thus far techniques are tested in artificially controlled settings. Typically the vision researcher has already determined the dataset's ...
 
Boundary preserving dense local regions
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Jaechul Kim,K. Grauman
Issue Date:June 2011
pp. 1553-1560
We propose a dense local region detector to extract features suitable for image matching and object recognition tasks. Whereas traditional local interest operators rely on repeatable structures that often cross object boundaries (e.g., corners, scale-space...
 
What's it going to cost you?: Predicting effort vs. informativeness for multi-label image annotations
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By S. Vijayanarasimhan, K. Grauman
Issue Date:June 2009
pp. 2262-2269
Active learning strategies can be useful when manual labeling effort is scarce, as they select the most informative examples to be annotated first. However, for visual category learning, the active selection problem is particularly complex: a single image ...
 
Discovering important people and objects for egocentric video summarization
Found in: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
By Yong Jae Lee,J. Ghosh,K. Grauman
Issue Date:June 2012
pp. 1346-1353
We developed an approach to summarize egocentric video. We introduced novel egocentric features to train a regressor that predicts important regions. Using the discovered important regions, our approach produces significantly more informative summaries tha...
 
Object-Graphs for Context-Aware Visual Category Discovery
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Yong Jae Lee,K. Grauman
Issue Date:February 2012
pp. 346-358
How can knowing about some categories help us to discover new ones in unlabeled images? Unsupervised visual category discovery is useful to mine for recurring objects without human supervision, but existing methods assume no prior information and thus tend...
 
Clues from the beaten path: Location estimation with bursty sequences of tourist photos
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Chao-Yeh Chen,K. Grauman
Issue Date:June 2011
pp. 1569-1576
Image-based location estimation methods typically recognize every photo independently, and their resulting reliance on strong visual feature matches makes them most suited for distinctive landmark scenes. We observe that when touring a city, people tend to...
 
Shape discovery from unlabeled image collections
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Yong Jae Lee, K. Grauman
Issue Date:June 2009
pp. 2254-2261
Can we discover common object shapes within unlabeled multi-category collections of images? While often a critical cue at the category-level, contour matches can be difficult to isolate reliably from edge clutter-even within labeled images from a known cla...
 
 1