Learning a classification model for segmentation

By Xiaofeng Ren, Jitendra Malik. 2003

1,942 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

Segment Anything (2023, background): “Segmentation is a broad field: there’s interactive segmentation [57, 109], edge detection [3], super pixelization [85], object proposal generation [2], foreground segmentation [94], semantic segmentation [90], instance segmentation [66], panoptic segmentation [59], etc.”
Semantic Image Segmentation via Deep Parsing Network (2015, background): “Existing works such as [31, 29, 9, 34, 11, 2, 8, 25, 22] can be generally categorized into two groups based on their definitions of the unary and pairwise terms of MRF.”
On the Binding Problem in Artificial Neural Networks (2020, background): “to learn a similarity function between segments (Ren and Malik, 2003; Endres and Hoiem, 2010; Kong and Fowlkes, 2018) or discrete graphical patterns (Lun et al.”
Playing for Data: Ground Truth from Computer Games (2016, background): “The identified patches are thus akin to superpixels [36], but have significant advantages over superpixels generated by a bottom-up segmentation algorithm.”
Geometric context from a single image (2005, background): “Each superpixel is assumed to correspond to a single label (superpixels have been shown to respect segment boundaries [24]).”
Multi-Scale Representation Learning on Proteins (2022, background): “Originally developed in computer vision [Ren and Malik, 2003, Mori et al., 2004, Kohli et al., 2009], superpixels are deﬁned as perceptually uniform regions in the image.”
GroupViT: Semantic Segmentation Emerges from Text Supervision (2022, methodology): “This pipeline has been successfully applied in image segmentation from superpixels [60], constructing region proposals for object detection [75, 94] and semantic segmentation [3].”
MILCut: A Sweeping Line Multiple Instance Learning Paradigm for Interactive Image Segmentation (2014, background): “Superpixels [31] have been proved effective in multiple vision tasks [22, 31].”
Adaptive pose priors for pictorial structures (2010, methodology): “We use publicly available code from [18], set to obtain about 125 superpixels per image.”
Robust Higher Order Potentials for Enforcing Label Consistency (2008, methodology): “A good example would be the method of [23] which uses inter and intra region similarity to measure the quality or goodness of a segment.”
Segmentation using superpixels: A bipartite graph partitioning approach (2012, background): “Over-segmentation occurs when image regions are segmented into smaller regions, each referred to as a “superpixel” [24].”
From contours to regions: An empirical evaluation (2009, background): “Many visual tasks can also benefit from the reduction in complexity achieved by transforming an image with millions of pixels into a few hundred or thousand " superpixels " [39].”
Improving Spatial Support for Objects via Multiple Segmentations (2007, methodology): “For the superpixel limit, we create superpixels [10] by over-segmenting each image into approximately 200 regions using Normalized Cuts and finding the optimal spatial support for each ground-truth object instance with respect to the superpixel map (example shown in Figure 3).”
BiSeNet V2: Bilateral Network with Guided Aggregation for Real-Time Semantic Segmentation (2020, methodology): “Traditional segmentation methods based on the threshold selection (Otsu, 1979), the region growing (Vincent and Soille, 1991), the super-pixel (Ren and Malik, 2003; Achanta et al., 2012; Van den Bergh et al., 2012) and the graph (Boykov and Jolly, 2001; Rother et al.”
SCALPEL: Segmentation Cascades with Localized Priors and Efficient Learning (2013, methodology): “Rather, our segmentation approach is more related to the greedy MCMC inference used by [17], again with the addition of shape priors and our modeling innovations.”
Shared Segmentation of Natural Scenes Using Dependent Pitman-Yor Processes (2008, methodology): “However, PY priors lead to efficient variational inference algorithms, avoiding the costly MCMC search required by other segmentation methods with region size priors [18, 19].”
Recovering Occlusion Boundaries from an Image (2011, methodology): “Some segmentation methods rely on 2D brightness, color, or texture cues to group the image pixels into perceptually similar regions (Shi and Malik 2000; Martin et al. 2001; Ren and Malik 2003; Felzenszwalb and Huttenlocher 2004; Arbelaez 2006).”
Shape-based pedestrian parsing (2011, methodology): “This approach is grounded in a long line of work starting with Ren and Malik [25] which scores multiple bottom-up segmentation hypotheses [21, 20, 26, 19, 5, 18].”