Local feature view clustering for 3D object recognition

By David G. Lowe. Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, 2001

599 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/9188b661f2ae65a080676617cc83b7d09773c59f

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

Total Recall: Automatic Query Expansion with a Generative Feature Model for Object Retrieval (2007, unknown): “[6], Lowe [8], Rothganger et al.”
Retrieving landmark and non-landmark images from community photo collections (2010, background): “We have used ﬂat k - means, where, as in [26], nearest cluster centers at each iteration have been found with the randomized k d-tree of Silpa-Anan and Hartley [29], and speciﬁcally using the FLANN library of Muja and Lowe [23].”
A sparse object category model for efficient learning and exhaustive recognition (2005, methodology): “A similar model, where the reference frame acts as a landmark is used by Lowe [14] and was studied in a probabilistic framework by Moreels et al.”
Cross-Dimensional Weighting for Aggregated Deep Convolutional Features (2015, background): “Triangulation embedding with democratic aggregation [17] was shown to give state-of-the-art results for SIFT-based architectures, while handling problems related to burstiness and interactions between unrelated descriptors prior to aggregation.”
Finding People in Images and Videos (2006, methodology): “…be based on points [Harris and Stephens 1988, Mikolajczyk and Schmid 2002], blobs (Laplacian of Gaussian [Lindeberg 1998] or Difference of Gaussian [Lowe 2001]), intensities [Kadir and Brady 2001, Ullman et al. 2001, Vidal-Naquet and Ullman 2003], gradients [Ronfard et al. 2002, Mikolajczyk et al.…”
Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection (2017, methodology): “Early approaches, such as [6], heavily depend on extracting local features such as SIFT [30], SURF [3], MSER [32] and matching them to retrieve instances [29, 48].”
Learning 6D Object Pose Estimation Using 3D Object Coordinates (2014, methodology): “For example, Lowe [15] used SIFT descriptors and clustered images from similar viewpoints into a single model.”
Depth-Encoded Hough Voting for Joint Object Detection and Shape Recovery (2010, background): “Other works propose to address the problem of detecting and estimating geometrical properties of single object instances [12,19,18,15]; while accurate pose estimation and 3D object reconstruction are demonstrated, these methods cannot be easily extended to incorporate intra-class variability so as t”
Video Google: a text retrieval approach to object matching in videos (2003, unknown): “Examples include [5] [6] [8] [11] [13] [12] [14] [16] [17].”
Structure from motion photogrammetry in physical geography (2016, methodology): “…(Morel and Yu, 2009), BRIEF (Calonder et al., 2010) and LDAHash (Strecha et al., 2012)), the Scale Invariant Feature Transform (SIFT) object recognition system is used most widely in SfM (Lowe, 1999, 2001, 2004) and has been shown by Lowe (2004) to perform well for changes in viewpoint of <40 .”
On the Use of SIFT Features for Face Authentication (2006, background): “Recently, theScale Invariant Feature Transform(SIFT) has emerged as a cut edge methodology in general object recognition as well as for other machine vision applications [13, 11, 12, 10, 5].”
Distinctive Image Features from Scale-Invariant Keypoints (2004, background): “For these reasons, 3D object recognition is best performed by integra ing features from multiple views, such as with local feature view clustering (Lowe, 2001).”
Pairwise Decomposition of Image Sequences for Active Multi-view Recognition (2016, background): “…viewpoints, a more stable set of features can be found by tracking those which are shared across multiple views and clustering images accordingly [23], or by learning their relative 2D displacements as the viewpoint changes, both with hard constraints for rigid bodies [17, 18] and ﬂexible…”