Detecting Object Boundaries Using Low-, Mid-, and High-level Information

By Songfeng Zheng, Alan L. Yuille, Zhuowen Tu. 2007

89 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Plant Science

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

In Plant Science

Citations in the corpus

Fast Edge Detection Using Structured Forests (2014, background): “F”
Auto-context and its application to high-level vision tasks (2008, methodology): “Compared to the algorithms using context [16, 8, 26 ], it learns an integrated model.”
Deep Colorization (2015, background): “It has recently been demonstrated that high-level understanding of an image is of great use for low-level vision problems ( e.g. image enhancement [18], edge detection [19]).”
Auto-Context and Its Application to High-Level Vision Tasks and 3D Brain Image Segmentation (2010, background): “We observe around 10-20 percent performance gain on a large MRI data set (> 500 images) over the baseline algorithm (no context); using the auto-context algorithm on top of all of these features gives further 5-10 percent improvement (the 20 percent performance gain is due to the complementariness…”
Multiscale Categorical Object Recognition Using Contour Fragments (2008, methodology): “An alternative method proposed in [ 56 ] is to learn to segment directly from the image.”
utomation of 3 D reconstruction of neural tissue from large volume of onventional serial section transmission electron micrographs (2008, background): “This approach is similar to machine earning-based boundary detection scheme proposed in (Martin et l., 2004; Zheng et al., 2007).”
Shape-Texture Debiased Neural Network Training (2020, background): “Moreover, researchers found that properly combining shape and texture can further recognition performance (Shotton et al., 2009; Zheng et al., 2007), demonstrating the superiority of possessing both features.”
Structured Forests for Fast Edge Detection (2013, background): “[37] combine low-, mid- and high-level cues and show improved results for object-specific edge detection.”
Sketch Tokens: A Learned Mid-level Representation for Contour and Object Detection (2013, methodology): “This was extended by [41] who used contextual and shape information to reﬁne the edge maps.”