Machine Learning
Automatic text detection and removal in video sequences
Pattern Recognition Letters
An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
Histograms of Oriented Gradients for Human Detection
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
A Comparison of Affine Region Detectors
International Journal of Computer Vision
Extraction of Text Objects in Video Documents: Recent Progress
DAS '08 Proceedings of the 2008 The Eighth IAPR International Workshop on Document Analysis Systems
Fast and robust text detection in images and video frames
Image and Vision Computing
Detecting and reading text in natural scenes
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
A Laplacian Approach to Multi-Oriented Text Detection in Video
IEEE Transactions on Pattern Analysis and Machine Intelligence
TILT: transform invariant low-rank textures
ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part III
Are MSER Features Really Interesting?
IEEE Transactions on Pattern Analysis and Machine Intelligence
ICDAR 2011 Robust Reading Competition Challenge 2: Reading Text in Scene Images
ICDAR '11 Proceedings of the 2011 International Conference on Document Analysis and Recognition
The estimation of the gradient of a density function, with applications in pattern recognition
IEEE Transactions on Information Theory
A Hybrid Approach to Detect and Localize Texts in Natural Scene Images
IEEE Transactions on Image Processing
| Hi-index | 0.00 |
Most objects with regular regions could be detected as Maximally Stable Extremal Regions (MSER) [20]. In this paper, We formulate object detection as a bi-label (object and non-object regions) segmentation problem, and propose a graph-based object detection method using edge-enhanced MSER. Specifically, we focus on detecting text in natural images, which is a special kind of object. First, edge-enhanced MSERs are detected as basic letter components; non-text MSERs are then efficiently eliminated by minimizing the cost function which combines both region-based and context-relevant information; and finally, mean-shift clustering is used to group text components into regions. The proposed method is naturally context-relevant, scale-insensitive and readily to be applied on detecting other objects. Experimental results on the ICDAR 2011 competition dataset show that the proposed approach outperforms state-of-the-art methods both in recall and precision.