PYRAMID METHODS IN IMAGE PROCESSING.

By Edward H. Adelson, P. Burt, Charles H. Anderson, J. M. Ogden, J. Bergen. 1984

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

Making Convolutional Networks Shift-Invariant Again (2019, methodology): “Classic hand-engineered computer vision and image processing representations, such as SIFT (Lowe, 1999), wavelets, and image pyramids (Adelson et al., 1984; Burt & Adelson, 1987) also extract features in a sliding window manner, often with some subsampling factor.”
Feature Pyramid Networks for Object Detection (2016, background): “Feature pyramids built upon image pyramids (for short we call these featur-ized image pyramids ) form the basis of a standard solution [1] (Fig.”
Multiview Transformers for Video Recognition (2022, background): ““Pyramid” structures [1] are one of the most popular multiscale representations for images and have been key in the early computer vision works, where their use has been widespread in multiple domains including feature descriptors [45], feature tracking [7, 46], image compression [9], etc.”
Beyond Gaussian Pyramid: Multi-skip Feature Stacking for action recognition (2014, background): “Multi-scale representation [1, 14] has been very popular for most image processing tasks such as image compression, image enhancement and object recognition.”
CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification (2021, background): “, image pyramids [1], scale-space representation [29], and coarse-to-fine approaches [28]).”
Scale-Aware Trident Networks for Object Detection (2019, methodology): “* Equal Contribution To remedy the large scale variation, an intuitive way is to leverage multi-scale image pyramids [1], which is popular in both hand-crafted feature based methods [12, 30] and current deep CNN based methods (Figure 1(a)).”
Optical Flow Estimation Using a Spatial Pyramid Network (2016, methodology): “We randomly scale images by a factor of [1, 2] and apply rotations at random within [−17◦, 17◦].”
Deep Feature Pyramid Reconfiguration for Object Detection (2018, background): “Detecting objects at vastly diﬀerent scales from images is a fundamental challenge in computer vision [1].”
Shiftable multiscale transforms (1992, background): “In order to create representations more suited for analysis of scale, many authors in signal processing and computer vision as well as biological vision have advocated transforms with constant logarithmic-width (constant Q) frequency bands [19], [31], [48], [24], [1], [34], [28].”
MAKING CONVOLUTIONAL NETWORKS SHIFT-INVARIANT AGAIN (2018, methodology): “Classic hand-engineered computer vision and image processing representations, such as SIFT (Lowe, 1999), wavelets, and image pyramids (Burt & Adelson, 1987; Adelson et al., 1984) also extract features in a sliding window manner, often with some subsampling factor.”
A Quantitative Analysis of Current Practices in Optical Flow Estimation and the Principles Behind Them (2013, background): “…to deal with large motions (Bergen et al. 1992; Brox et al. 2004), texture decomposition (Wedel et al. 2008a,b) or high-order ﬁlter constancy (Adelson et al. 1984; Brox et al. 2004; Glaer et al. 1983; Lempitsky et al. 2010; Zimmer et al. 2009) to reduce the inﬂuence of lighting changes,…”
D2-Net: A Trainable CNN for Joint Description and Detection of Local Features (2019, methodology): “In order to obtain features that are more robust to scale changes, we propose to use an image pyramid [2], as typically done in hand-crafted local feature detectors [27,29,31] or even for some object detectors [16].”
Zoom In and Out: A Mixed-scale Triplet Network for Camouflaged Object Detection (2022, methodology): “Its ideas have been widely used in computer vision, including the image pyramid [2] and the feature pyramid [22].”
MUSIQ: Multi-scale Image Quality Transformer (2021, background): “Another aspect for improving IQA models is to imitate the human visual system which captures an image in a multi-scale fashion [1].”
Learning Feature Pyramids for Human Pose Estimation (2017, background): “, HOG, on different levels of image pyramids [1, 16].”
Person Re-identification by Deep Learning Multi-scale Representations (2017, background): “In general, object/event/scene representation for recognition at explicitly different scales is widely adopted in computer vision [24, 43, 10], in particular the idea of constructing feature pyramids from image pyramid inputs [2, 24, 34].”