Packing bag-of-features

By Herv'e J'egou, Matthijs Douze, Cordelia Schmid. 2009

195 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Aggregating Local Image Descriptors into Compact Codes (2012, methodology): “A low dimensional vector can be indexed efﬁciently, but its discriminative power may be too low.”
Product Quantization for Nearest Neighbor Search (2011, background): “This memory cost can be reduced further using index compression [25], [26], which may reduce the average cost of storing the identifier to about 8 bits, depending on parameters.”
Image Retrieval for Image-Based Localization Revisited (2012, background): “In contrast, the inverted file indices and feature geometry needed for image retrieval can be stored very efficiently [15, 24].”
Visual Place Recognition with Repetitive Structures (2015, unknown): “In addition, the inverted files can be compressed (without a loss of performance) which further reduces the required memory, typically by a factor of 4 [56].”
Searching with quantization: approximate nearest neighbor search using short codes and distance estimators (2009, background): “This memory cost can be further reduced by the use of index compression [26, 27], which may reduce the average cost of storing the identifier to about 8 bits, depending on parameters(2).”
Empowering Visual Categorization With the GPU (2011, background): “The same argument applies to other indexing structures such as mini bag-of-features (miniBOF) [31]: accuracy is sacrificed in return for faster computation.”
High-dimensional signature compression for large-scale image classification (2011, background): “The aggregation process may be unbalanced [26, 33] or balanced [12], i.”
Image retrieval with geometry-preserving visual phrases (2011, methodology): “Spatial verification methods [15] and query expansion [5] have been proposed for re-ranking at the post-processing step, and many methods [9, 22, 8 , 14] have been introduced to decrease the memory usage for the inverted files.”
Sketch-based manga retrieval using manga109 dataset (2015, methodology): “This kind of task has been tackled in spatial veriﬁcation-based reranking methods [25], [26], [27], [28], [29], [30], [31].”
Learning Representations by Predicting Bags of Visual Words (2020, methodology): “In our experiments we use the binary version (4) [31, 56] for ImageNet and the histogram version (3) for CIFAR-100 and MiniImageNet.”
Aggregating local descriptors into a compact image representation (2010, unknown): “This timing is better than the one reported in [ 9 ], and this for a significantly better accuracy.”
Large-scale image retrieval with compressed Fisher vectors (2010, methodology): “We now compare in Figure 4 our results with the recent results of [ 13 ] based on compressed BOV vectors.”
Visual landmark recognition from Internet photo collections: A large-scale evaluation (2014, methodology): “roaches in this evaluation. 3.4. Eliminating Redundancy Several methods have been proposed to reduce the size of the visual search index. An obvious method is to apply standard compression techniques [40], which reduces memory consumption at the cost of computational eciency. Instead, we are interes”