LDAHash: Improved Matching with Smaller Descriptors

By Christoph Strecha, A. Bronstein, Michael M. Bronstein, Pascal Fua. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012

632 citations recorded by the source corpus.

SIFT-like local feature descriptors are ubiquitously employed in computer vision applications such as content-based retrieval, video analysis, copy detection, object recognition, photo tourism, and 3D reconstruction. Feature descriptors can be designed to be invariant to certain classes of photometric and geometric transformations, in particular, affine and intensity scale transformations. However, real transformations that an image can undergo can only be approximately modeled in this way, and thus most descriptors are only approximately invariant in practice. Second, descriptors are usually high dimensional (e.g., SIFT is represented as a 128-dimensional vector). In large-scale retrieval and matching problems, this can pose challenges in storing and retrieving descriptor data. We map the descriptor vectors into the Hamming space in which the Hamming metric is used to compare the resulting representations. This way, we reduce the size of the descriptors by representing them as short binary strings and learn descriptor invariance from examples. We show extensive experimental validation, demonstrating the advantage of the proposed approach.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/5bba402f7dd73a8c2233d069d84fe91247fdbf55

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

Fast search in Hamming space with multi-index hashing (2012, methodology): “, [5, 12, 18, 21, 24]), while others focus on semantic similarity (e.”
Fast Exact Search in Hamming Space With Multi-Index Hashing (2013, neutral): “Nearest neighbor (NN) search on binary codes is used for image search [29], [35], [38], matching local features [2], [7], [16], [33], image classiﬁcation [6], object segmentation [19], and parameter estimation [32].”
Iterative quantization: A procrustean approach to learning binary codes (2011, background): “Recently, the vision community has devoted a lot of attention to the problem of learning similarity-preserving binary codes for representing large-scale image collections [12, 16, 18, 19, 20, 21].”
Learning descriptors for object recognition and 3D pose estimation (2015, methodology): “Some of these methods are applied to existing descriptors to make them more discriminative, such as in [10, 31], but others are trained directly on image data.”
Shape google: Geometric words and expressions for invariant shape retrieval (2011, methodology): “In [Strecha et al. 2010], similarity-sensitive hashing algorithms were applied to local SIFT descriptors to improve the performance of feature matching in wide-baseline stereo reconstruction problems.”
Large-Scale Supervised Multimodal Hashing with Semantic Correlation Maximization (2014, background): “…and Indyk 2008; Kulis and Darrell 2009; Lin, Ross, and Yagnik 2010; Wang, Kumar, and Chang 2010; Norouzi and Fleet 2011; Kong, Li, and Guo 2012; Heo et al. 2012; Liu et al. 2012a; Kong and Li 2012b; Norouzi, Fleet, and Salakhutdinov 2012; Strecha et al. 2012; Ge et al. 2013; Neyshabur et al. 2013).”
Supervised hashing with kernels (2012, methodology): “We uniformly randomly sample 100 and 200 images from each class respectively, constituting 1K and 2K labeled subsets for training (semi-)supervised hashing methods SSH, LDAH, BRE, MLH, KSH 0 and KSH.”
Worldwide Pose Estimation Using 3D Point Clouds (2012, methodology): “In the future, this could be improved through the use of more compact descriptors [32], or by intelligently compressing the database.”
Understanding Locally Competitive Networks (2014, methodology): “ﬁcient retrieval of data based on high level similarity even though no pair-wise loss was used during training. We compare to DiffHash, a supervised similarity-preserving hashing approach proposed by Strecha et al. (2012), trained on the non-binarized features from the network. Supervision is repres”
Pose Estimation for Augmented Reality: A Hands-On Survey (2016, neutral): “Since feature descriptors from the methods above have floating-point values, they can be compacted into a binary string with machine learning techniques [128], [135].”
Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “For example, the LDA technique is utilized in Strecha et al. (2012) to aid hashing.”
Learning Hash Functions Using Column Generation (2013, background): “, 2010b), local descriptor compression (Strecha et al., 2011), image matching (Korman & Avidan, 2011), and many more.”
Hashing for Distributed Data (2015, background): “Introduction Large scale similarity search is a fundamental building block in many machine learning applications such as clustering, classification and matching (Strecha et al., 2012; Cheng et al., 2014).”
Linear cross-modal hashing for efficient multimedia search (2013, methodology): “For example, supervised hashing methods [14, 17, 19, 24, 28] improve the hashing performance using the pre-provided pairs of data with the assumption that there is “similar” or “dissimilar” pairs in datasets.”
Optimized Cartesian K-Means (2014, neutral): “C V ] 16 M ay 2 01 4 Index Terms—Clustering, Cartesian product, Nearest neighbor search ✦”
Assessing the Accuracy of Georeferenced Point Clouds Produced via Multi-View Stereopsis from Unmanned Aerial Vehicle (UAV) Imagery (2012, methodology): “A number of alternatives to SIFT exist such as Gradient Location and Orientation Histogram (GLOH) [15], Speeded Up Robust Features (SURF) [16], LDAHash [17] and Principal Component Analysis (PCA)SIFT [18].”