Large displacement optical flow

By Thomas Brox, C. Bregler, Jitendra Malik. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009

365 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/aa9f82781a0d844faec8616ffbe68113536fefea

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Large Displacement Optical Flow: Descriptor Matching in Variational Motion Estimation (2011, positive): “The present paper extends a preliminary conference version [10], where region correspondences serve to recover large displacements.”
Optical Flow Estimation Using a Spatial Pyramid Network (2016, background): “Note that pyramids have well-known limitations for dealing with large motions [8, 34].”
TAPIR: Tracking Any Point with per-frame Initialization and temporal Refinement (2023, neutral): “Optical Flow focuses on dense motion estimation between image pairs [20, 34]; methods can be divided into classical variational approaches [4, 5] and deep learning approaches [14, 21, 43, 55, 57, 66].”
Dense Point Trajectories by GPU-Accelerated Large Displacement Optical Flow (2010, methodology): “In the tennis sequence [11], there are frames”
Self-supervised Video Representation Learning for Correspondence Flow (2019, positive): “Later works integrate feature matching to initialize sparse matching, and then interpolate into dense flow maps in a pyramidal coarse-to-fine manner [5, 33, 45].”
Learning Correspondence From the Cycle-Consistency of Time (2019, background): “For example, Brox and Malik [5] introduced a descriptor that matches region hierarchies and provides dense and subpixel-level estimation of ﬂow.”
Visual attribute transfer through deep image analogy (2017, methodology): “Some methods combine sparse features with dense matching to cope with large-displacement optical flow [Brox et al. 2009], while others perform matching of visually different scenes (e.”
From contours to regions: An empirical evaluation (2009, background): “Our generic grouping machinery has found use in optical flow [8] and object recognition [19] applications.”
Contour Detection and Hierarchical Image Segmentation (2011, methodology): “In particular, our gPb-owt-ucm segmentation algorithm has found use in optical flow [14] and object recognition [15], [16] applications.”
A Database and Evaluation Methodology for Optical Flow (2007, methodology): “Such sparse matching method can be combined with the continuous energy minimization approaches in a variety of ways ( Brox et al. 2009; Liu et al. 2008 ;R en2008; Xu et al. 2008).”
Dense Semantic Correspondence Where Every Pixel is a Classifier (2015, neutral): “More recently, a number of variational methods have used sparse descriptor matching to anchor larger displacements [2, 20].”
The Generalized PatchMatch Correspondence Algorithm (2010, background): “More advanced methods [6, 7] that start with sparse matches and then propagate them densely suffer from similar problems.”
Generalized Boundaries from Multiple Image Interpretations (2012, methodology): “We combine the low-level color and depth input with large-displacement optical ﬂow [2], which is useful for large inter-frame body movements.”
Motion Detail Preserving Optical Flow Estimation (2012, neutral): “Brox et al. [9], in computing large-displacement optical flow, pointed out that if flow structure is smaller than its displacement, the latter may not be well estimated.”
Joint SFM and detection cues for monocular 3D localization in road scenes (2015, methodology): “While more accurate optical flow methods are available [1, 21], we choose TV-L1 for its balance of accuracy and speed.”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, methodology): “In this work, we are interested in a new, higher level of image alignment: aligning two images from different 3D scenes but sharing similar scene characteristics.”
Non-rigid dense correspondence with applications for image enhancement (2011, methodology): “More advanced methods combine sparse features with dense matching to cope with large-displacement optical ﬂow [Brox et al. 2009], and non-rigid matching of highly different scenes [Liu et al. 2008a].”
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (2020, methodology): “Improvements have been made by defining better matching costs [45,10] and regularization terms [38].”