An Iterative Image Registration Technique with an Application to Stereo Vision

By B. D. Lucas, Takeo Kanade. 1981

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

Perceiver IO: A General Architecture for Structured Inputs & Outputs (2021, background): “Optical flow is a decades-old open problem in computer vision [49, 33].”
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.”
One-shot-learning gesture recognition using HOG-HOF features (2013, methodology): “ is the Histogram of Oriented Gradients, proposed by Dalal and Triggs (2005). A good method for capturing the size and direction of motion is computing the Optical Flow using the Lucas-Kanade method (Kanade and Lucas, 1981; Lucas, 1984) and creating a histogram of ow. Motivation for these choices is”
Learned Video Compression (2018, methodology): “The problem of optical ﬂow estimation has been widely studied over the years, with thousands of solutions developed using tools from partial differential equations [15, 24, 10, 11, ... ] to, more recently, machine learning [45, 9, 17, 32, ...].”
Unsupervised Learning of Edges (2015, background): “Optical Flow: The estimation of optical flow is a classic problem in computer vision [18, 24].”
Video-to-Video Synthesis (2018, background): “If the optical flow [46] between consecutive frames is known, we can estimate the next frame by warping the current frame [54, 70].”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, background): “Image alignment becomes even more difficult in the object recognition scenario, where the goal is to align different instances of the same object category, as illustrated in Figure 1 (b).”
Robustness via Retrying: Closed-Loop Robotic Manipulation with Self-Supervised Learning (2018, background): “It is possible to use off-the-shelf trackers [30, 31, 32, 33] to mitigate this issue.”
Solving Linear Inverse Problems Using the Prior Implicit in a Denoiser (2020, methodology): “Since the family of observation densities, p σ ( y ) forms a scale-space representation of p ( x ) , the algorithm may be viewed as an adaptive form of coarse-to-ﬁne optimization [43–46].”
A Biologically Inspired System for Action Recognition (2007, methodology): “Optical flow based S1 units: A second set of S1 units was obtained by computing the optical flow of the input image sequence using Lucas & Kanade’s algorithm [14].”
Deep Global Registration (2020, methodology): “To maximize the accuracy and speed of 3D registration, researchers have developed geometric feature descriptors [22, 11, 41, 9], pose optimization algorithms [38, 45, 24, 50], and end-to-end feature learning and registration pipelines [41, 2].”
PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking (2023, background): “Early motion estimation methods cast point tracking as an optimization problem defined on handcrafted features [27, 37, 57, 52, 12], and these techniques continue to drive structure-from-motion [11, 31, 43] and simultaneous localization and mapping systems [54].”
A latent model of discriminative aspect (2009, methodology): “Given the bounding box of the actor and the silhouette, the optic flow is computed using Lucas-Kanade algorithm [ 21 ].”
COTR: Correspondence Transformer for Matching Across Images (2021, methodology): “Dense methods, by contrast, usually model small temporal changes, such as optical ﬂow in video sequences, and rely on local smoothness [35, 24].”
Motion-based motion deblurring (2004, methodology): “We compute the motion between successive frames using a multiresolution iterative algorithm that minimizes the following optical flow based error function [ 22 ]:”
GPU-based Video Feature Tracking And Matching (2006, methodology): “The KLT tracking algorithm [6,7] computes displacement of features or interest points between consecutive video frames when the image brightness constancy constraint is satisfied and image motion is fairly small.”
GVINS: Tightly Coupled GNSS–Visual–Inertial Fusion for Smooth and Consistent State Estimation (2021, methodology): “The strong corners [29] within the image are detected as feature points and are further tracked by the iterative Lucas-Kanade method [30].”