OBJ CUT

By M. Pawan Kumar, Philip H. S. Torr, Andrew Zisserman. 2005

413 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Convolutional feature masking for joint object and stuff segmentation (2014, background): “Semantic segmentation [14, 19, 24, 2] aims to label each image pixel to a semantic category.”
Perceiver: General Perception with Iterative Attention (2021, background): “Re-entrant processing of this kind (sometimes referred to as top-down processing) has a long history in computer vision (Borenstein et al., 2004; Kumar et al., 2005; Carreira et al., 2016; Hu & Ramanan, 2016; Yang et al., 2018; Lin et al., 2020).”
UberNet: Training a Universal Convolutional Neural Network for Low-, Mid-, and High-Level Vision Using Diverse Datasets and Limited Memory (2016, background): “for integrating segmentation and recognition [9, 47, 50, 53, 68, 72, 99], and we believe that successfully addressing this it is imperative to have a single network that can succesfully handle all of the involved tasks.”
Monocular Object Instance Segmentation and Depth Ordering with CNNs (2015, neutral): “Combination of detection and segmentation has been the most common approach to instance segmentation: once an object is localized with a bounding box, top-down information such as shape or appearance can be used to carve out the object within the box [20].”
Pixel-Level Encoding and Depth Layering for Instance-Level Semantic Labeling (2016, methodology): “Instead of bounding boxes, [18] uses a layered pictorial structure (LPS) model, where shape exemplars for object parts are mapped to the image in a probabilistic way.”
Instance-Level Segmentation for Autonomous Driving with Deep Densely Connected MRFs (2015, methodology): “The most common approach to object instance segmentation is to first localize objects with a set of bounding boxes, and then exploit top-down information such as object’s shape and appearance in order to accurately segment the object within each box [17].”
Recognition using regions (2009, methodology): “It has been addressed in the past by techniques such as OBJ CUT [17].”
Amodal Instance Segmentation (2016, methodology): “Early methods [4,37,25,23,26] investigated ways of combining top-down and bottom-up segmentation approaches and incorporating class-speciﬁc or object-speciﬁc templates of the foreground object’s appearance.”
The Shape Boltzmann Machine: A Strong Model of Object Shape (2012, methodology): “Non-parametric approaches employ what is effectively a large database of template shapes [12] or shape fragments [3, 15].”
Augmenting deformable part models with irregular-shaped object patches (2012, methodology): “[14], combines CRFs with Pictorial Structures to solve a joint segmentation-detection problem.”
ClassCut for Unsupervised Class Segmentation (2010, methodology): “We discussed in the introduction that our method employs energy minimization techniques used in interactive segmentation [1,3,4,14,15], and how it is related to supervised [5,7,19] as well as to unsupervised [2,10–12] class segmentation methods.”
Star Shape Prior for Graph-Cut Image Segmentation (2008, methodology): “The advantage of using a generic star shape prior is that it can be directly incorporated in the optimization procedure, no expensive registration between the model and the image, like in [26, 27] is required.”
Instance-Level Segmentation with Deep Densely Connected MRFs (2015, methodology): “The most common approach to object instance segmentation is to first localize objects with a set of bounding boxes, and then exploit top-down information such as object’s shape and appearance in order to accurately segment the object within each box [14].”
Semantic segmentation using regions and parts (2012, background): “, [31, 20, 1, 27, 15, 10, 37, 5, 12, 21]; the second one is addressed naturally in a multiscale scanning window detec-”
Object Recognition by Sequential Figure-Ground Ranking (2011, methodology): “One way to improve such results is to integrate low-level information as image edges (Kumar et al. 2005; Toshev et al. 2010) or bottom-up hierarchical segmentations (Borenstein and Ullman 2008).”
TextonBoost: Joint Appearance, Shape and Context Modeling for Multi-class Object Recognition and Segmentation (2006, background): “Joint detection and segmentation of a single object class has been achieved by several authors [7, 8, 9].”