Object Recognition by Sequential Figure-Ground Ranking
International Journal of Computer Vision
Bottom-up perceptual organization of images into object part hypotheses
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part I
Automatic segmentation of unknown objects, with application to baggage security
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part II
Semantic segmentation with second-order pooling
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part VII
Classification of segmented objects through a multi-net approach
ANNPR'12 Proceedings of the 5th INNS IAPR TC 3 GIRPR conference on Artificial Neural Networks in Pattern Recognition
Using models of objects with deformable parts for joint categorization and segmentation of objects
ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part II
Multiscale Symmetric Part Detection and Grouping
International Journal of Computer Vision
Break and conquer: efficient correlation clustering for image segmentation
SIMBAD'13 Proceedings of the Second international conference on Similarity-Based Pattern Recognition
Hierarchical object discovery and dense modelling from motion cues in RGB-D video
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
Probabilistic Joint Image Segmentation and Labeling by Figure-Ground Composition
International Journal of Computer Vision
| Hi-index | 0.14 |
We present a novel framework to generate and rank plausible hypotheses for the spatial extent of objects in images using bottom-up computational processes and mid-level selection cues. The object hypotheses are represented as figure-ground segmentations, and are extracted automatically, without prior knowledge of the properties of individual object classes, by solving a sequence of Constrained Parametric Min-Cut problems (CPMC) on a regular image grid. In a subsequent step, we learn to rank the corresponding segments by training a continuous model to predict how likely they are to exhibit real-world regularities (expressed as putative overlap with ground truth) based on their mid-level region properties, then diversify the estimated overlap score using maximum marginal relevance measures. We show that this algorithm significantly outperforms the state of the art for low-level segmentation in the VOC 2009 and 2010 data sets. In our companion papers [1], [2], we show that the algorithm can be used, successfully, in a segmentation-based visual object category recognition pipeline. This architecture ranked first in the VOC2009 and VOC2010 image segmentation and labeling challenges.