SIFT Flow: Dense Correspondence across Different Scenes

By Ce Liu, Jenny Yuen, Antonio Torralba, Josef Sivic, William T. Freeman. 2008

748 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

Learning to generate chairs with convolutional neural networks (2014, methodology): “In Table 1 we show the performance of our algorithm compared to human performance and two baselines: SIFT ﬂow [21] and Deformable Spatial Pyramid [15] (DSP).”
A Multi-View Embedding Space for Modeling Internet Images, Tags, and Their Semantics (2012, background): “One of these is nonparametric image parsing (Liu et al., 2010; Tighe and Lazebnik, 2010) where, given a query image, a small number of similar training images is retrieved and labels are transferred from these images to the query.”
Depth Map Prediction from a Single Image using a Multi-Scale Deep Network (2014, methodology): “Karsch et al. [?] use a kNN transfer mechanism based on SIFT Flow [?] to estimate depths of static backgrounds from single images, which they augment with motion information to better estimate moving foreground subjects in videos.”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, neutral): “8.”
Plant Species Identification Using Computer Vision Techniques: A Systematic Literature Review (2017, background): “But DSIFT is not scale-invariant, to make it adaptable to changes in scale, local features are sampled by different scale patches within an image [84].”
Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (2024, background): “Early works [23, 36, 50] primarily relied on handcrafted features and traditional computer vision techniques.”
3DNN: 3D Nearest Neighbor (2014, methodology): “These non-parametric approaches have been shown to achieve amazing performance for a wide variety of complex computer vision and graphics tasks ranging (Oliva and Torralba 2001), to motion synthesis (Liu et al. 2008) and even image localization (Hays and Efros 2008).”
InLoc: Indoor Visual Localization with Dense Matching and View Synthesis (2019, methodology): “Instead of adjusting the parameters of handcrafted features [47], [49] (patch sizes, strides, scales) to match across images with significant viewpoint changes, we use an image representation extracted by a CNN architecture as”
Monocular Fisheye Camera Depth Estimation Using Sparse LiDAR Supervision (2018, methodology): “[11] recommended a k-Nearest-Neighbor (kNN) transfer mechanism which can achieve better alignment which hinges on SIFT Flow [12] to estimate depths from single images of static backgrounds.”
Non-rigid dense correspondence with applications for image enhancement (2011, methodology): “Therefore, we adopted the metric used by [Liu et al. 2008a].”
Regenerative morphing (2010, methodology): “9) - simple blending, SIFT-flow [12]+blending, and our method (right).”
A survey on activity recognition and behavior understanding in video surveillance (2013, background): “[38] optical flow on pixel resolution o c s s improved speed/accuracy, detect occlusion between objects [67] gradient-based + FPGA i c s s investigation of top’s spinning and a human’s pitching motion [18] frame straddling + FPGA i,o c s s less computation cost, high speed, less delay involve [97] c”
Beyond pixels: exploring new representations and applications for motion analysis (2009, methodology): “As reported in [69], the computation time for 145 x 105 images with an 80 x 80 searching neighborhood is 50 seconds.”
Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Large Displacement Optical Flow: Descriptor Matching in Variational Motion Estimation (2011, neutral): “In contrast, Liu et al. [23] have recently proposed a method that computes dense correspondence fields between two different scenes.”
Nonparametric Scene Parsing via Label Transfer (2011, background): “This paper is organized as follows: In Section 2, we briefly survey the object recognition and detection literature.”
Semantic Object Selection (2014, methodology): “Positive exemplar database: Objects on white background and exemplars from PASCAL VOC (last 2 columns). object proposal using SIFT flow based image warping [16] to produce a location prior.”