SiftGPU : A GPU Implementation of Scale Invariant Feature Transform (SIFT)

By Changchang Wu. 2007

537 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/288020dbbecf002fa15d3db254a39681e954515e

How researchers see this paper

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

Citations in the corpus

A Survey on Structure from Motion (2017, methodology): “Additionally, by the integration of the GPU implementation of SIFT given in [84], VisualSfM can be used as an end-to-end SfM engine taking as input a set of images and producing as output a motion and a 3D structure estimate (however, such a black box usage may result in degradation of accuracy in…”
Image Retrieval for Image-Based Localization Revisited (2012, methodology): “An SfM reconstruction of the scene was obtained using the freely available Bundler software [31, 39] and SiftGPU [38].”
RGB-D Mapping: Using Depth Cameras for Dense 3D Modeling of Indoor Environments (2010, background): “These steps can be implemented with arbitrary visual features (in our implementation, we compute SIFT features using SIFTGPU [32]).”
RGB-D mapping: Using Kinect-style depth cameras for dense 3D modeling of indoor environments (2012, methodology): “In the conference version of this paper (Henry et al. 2010), we used SIFT features computed with SIFTGPU (Wu 2007).”
3-D Mapping With an RGB-D Camera (2014, methodology): “We use the OpenCV implementations in our implementation, except for SIFT, where we employ a GPU-based implementation [42].”
The use of close-range photogrammetry in zooarchaeology: Creating accurate 3D models of wolf crania to study dog domestication (2016, methodology): “…perform keypoint detection (unambiguous feature-points in each image) through SIFT algorithms (i.e. Scale Invariant Feature Transforms; Lowe, 1999; Wu, 2007), tie-points matching (matching points between pairs of photographs), sparse 3D points generation (Sholts et al., 2011;Wu, 2013;Wu et al.,…”
Manipulator and object tracking for in-hand 3D object modeling (2011, methodology): “In our implementation, we use SiftGPU (Wu, 2007) for feature detection.”
Robust monocular SLAM in dynamic environments (2013, methodology): “Different to PTAM, we extract SIFT features for each online frame and match them between adjacent frames, where SiftGPU [39] is employed in our implementation.”
Parallel search of k-nearest neighbors with synchronous operations (2012, unknown): “It has parallel implementation on GPUs [33].”
The MOPED framework: Object recognition and pose estimation for manipulation (2011, methodology): “MOPED uses SIFTGPU (Wu 2007) as its main feature extraction algorithm.”
An evaluation of the RGB-D SLAM system (2012, methodology): “Since SIFT features are computationally much more demanding than SURF and ORB, we also make use of a GPU based implementation of SIFT [32].”
GMS: Grid-Based Motion Statistics for Fast, Ultra-Robust Feature Correspondence (2017, background): “Many of these works have GPU-acceleration [45, 40, 7] permitting real (or near-real) time performance.”