Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Large Displacement Optical Flow: Descriptor Matching in Variational Motion Estimation

By Thomas Brox, Jitendra Malik. 2011

1,500 citations recorded by the source corpus.

Subfields: Aerospace Engineering, 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 Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Delving Deeper into Convolutional Networks for Learning Video Representations (2015, background): “The motion of video patches tend to be restricted to a local neighborhood (Brox & Malik, 2011).”
C3D: Generic Features for Video Analysis (2014, methodology): “However, this method uses Brox’s optical ﬂows [4] as low level input signals.”
FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks (2016, methodology): “The original model in Ochs et al. [14] was built on Large Displacement Optical Flow [7].”
FlowNet: Learning Optical Flow with Convolutional Networks (2015, methodology): “The recent focus was on large displacements, and combinatorial matching has been integrated into the variational approach [6, 35].”
Unsupervised Learning of Edges (2015, methodology): “side of our scope, instead, our work is most closely related to methods that leverage sparse matches or image edges for flow estimation [6, 42, 31].”
Playing for Benchmarks (2017, methodology): “We benchmark the performance of four well-known optical ﬂow algorithms: LDOF [7], EpicFlow [51], Flow-Fields [3], and FlowNet [14].”
On the Link between Gaussian Homotopy Continuation and Convex Envelopes (2015, methodology): “Just within the past few years, it has been utilized for lowrank matrix recovery [45], error correction [48], super resolution [19], photometric stereo [72], image segmentation [35], face alignment [57], 3D surface estimation [1], motion estimation in videos [61], optical flow [10,62], shape and ill”
Deep Discrete Flow (2016, background): “Popular similarity measures for optical flow include brightness and gradient constancy [24], normalized cross-correlation [12], SIFT [30], Daisy [11] and hierarchical histograms of oriented gradients [32].”
Layered RGBD scene flow estimation (2015, background): “Some notable variants include image-dependent, non-local motion priors to recover sharp motion boundaries [38, 47, 51], fusion moves to reach stronger local minima [24], embedding feature matching to track small, fast-moving objects [10, 45, 49], exploiting the epipolar constraint for egocentric sce”
Learning to Predict Gaze in Egocentric Video (2013, methodology): “Due to the substantial motion in egocentric videos, we apply Large Displacement Optical Flow (LDOF) [4] between two consecutive frames to get the motion field.”
Flow Fields: Dense Correspondence Fields for Highly Accurate Large Displacement Optical Flow Estimation (2015, background): “One of the ﬁrst works that integrated sparse descriptor matching for improved large displacement performance was Brox and Malik [7].”
Deep Feature Flow for Video Recognition (2016, background): “The recent focus is on large displacements [3], and combinatorial matching (e.”
An Unsupervised Learning Model for Deformable Medical Image Registration (2018, background): “Extensions that better handle large displacements or dramatic changes in appearance include feature-based matching [9, 28] and nearest neighbor fields [10].”
Learning to Extract Motion from Videos in Convolutional Neural Networks (2016, methodology): “In particular, we do not perform explicit feature tracking and long-range matching, which are the highlights of the best-performing methods on this dataset ( e.g. [39, 4]).”
Predicting the Future from First Person (Egocentric) Vision: A Survey (2021, methodology): “The most common approaches are TVL1 [125], FlowNet 2.0 [126], LDOF [127], the Horn-Schunck algorithm [121], and SMD (Salient Motion Detection) [123].”