Fisher Kernels on Visual Vocabularies for Image Categorization

Perronnin, Florent; Dance, Christopher R.

doi:10.1109/cvpr.2007.383266

Cited by 1,376 publications

(1,074 citation statements)

References 9 publications

Supporting

Mentioning

1,055

Contrasting

Unclassified

Order By: Relevance

“…This approach is an approximation of the original method from Perronnin et al [33] who creates a large vector for each image: one part being related to a global ("universal" in [33]) visual dictionary based on an unsupervised GMM trained on all data, and the other part is related to a supervised (category) visual dictionary using Gaussian Mixture Models trained on each category. Inspired by these works, [20] Jegou et al proposed a simpler approach in which K-Means algorithm applied to VLAD descriptors approximates the universal visual dictionary and gets performance comparable to Perronnin et al ones.…”

Section: Combined Supervised and Unsupervised Visual Dictionariesmentioning

confidence: 99%

“…We have also carried out experiments using the state of the art approach: Vectors of Locally Aggregated Descriptors (VLAD) [20], which is derived from the method based on bag of features and Fisher Kernels proposed in [33]. VLAD approach aggregates keypoint descriptors into a single vector in a more efficient way than bag of features.…”

Section: Unsupervised Visual Dictionarymentioning

confidence: 99%

“…Finally, in order to compare somewhat fairly our approach to the current best method on PASCAL Challenge [33], we have carried out experiments using the concatenation of all our supervised features and the global dictionary based on VLAD descriptors. This approach is an approximation of the original method from Perronnin et al [33] who creates a large vector for each image: one part being related to a global ("universal" in [33]) visual dictionary based on an unsupervised GMM trained on all data, and the other part is related to a supervised (category) visual dictionary using Gaussian Mixture Models trained on each category.…”

Section: Combined Supervised and Unsupervised Visual Dictionariesmentioning

confidence: 99%

“…Then, a kernel function, RBF or Fisher kernel, is considered. Such strategies proved to be very powerful in a batch learning context, when a large training set is available [20,33,34]. In [32], Perronnin first proposed to build two kinds of visual dictionaries per category: one "universal" visual dictionary is build on the whole database with a Gaussian mixture model (GMM) without any supervision, while the other one dedicated to the category uses labelled training data to learn the GMM of that category.…”

Section: Introductionmentioning

confidence: 99%

“…In [32], Perronnin first proposed to build two kinds of visual dictionaries per category: one "universal" visual dictionary is build on the whole database with a Gaussian mixture model (GMM) without any supervision, while the other one dedicated to the category uses labelled training data to learn the GMM of that category. In [33], Perronnin et al applied Fisher kernel on visual words modeled with a combination of "universal" and "categorical" GMMs, restricted to diagonal variance matrices for each component of the mixture. Deriving a diagonal approximation of the Fisher matrix of a GMM, they obtain a (2d + 1) × k − 1 dimensional vector representation of an image feature set, or d × k dimensional vector when considering only the components associated with either the means or the variances of the GMM.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Incremental kernel learning for active image retrieval without global dictionaries

Gosselin

Precioso

Philipp‐Foliguet

2011

Pattern Recognition

View full text Add to dashboard Cite

In content-based image retrieval context, a classic strategy consists in computing off-line a dictionary of visual features. This visual dictionary is then used to provide a new representation of the data which should ease any task of classification or retrieval. This strategy, based on past research works in text retrieval, is suitable for the context of batch learning, when a large training set can be built either by using a strong prior knowledge of data semantics (like for textual data) or with an expensive off-line pre-computation. Such an approach has major drawbacks in the context of interactive retrieval, where the user iteratively builds the training data set in a semi-supervised approach by providing positive and negative annotations to the system in the relevance feedback loop. The training set is thus built for each retrieval session without any prior knowledge about the concepts of interest for this session. We propose a completely different approach to build the dictionary on-line from features extracted in relevant images. We design the corresponding kernel function, which is learnt during the retrieval session. For each new label, the kernel function is updated with a complexity linear with respect to the size of the database. We propose an efficient active learning strategy for the weakly supervised retrieval method developed in this paper. Moreover this framework allows the combination of features of different types. Experiments are carried out on standard databases, and show that a small dictionary can be dynamically extracted from the features with better performances than a global one.

show abstract

Section: Combined Supervised and Unsupervised Visual Dictionariesmentioning

confidence: 99%

Section: Unsupervised Visual Dictionarymentioning

confidence: 99%

Section: Combined Supervised and Unsupervised Visual Dictionariesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Incremental kernel learning for active image retrieval without global dictionaries

Gosselin

Precioso

Philipp‐Foliguet

2011

Pattern Recognition

View full text Add to dashboard Cite

show abstract

Detection of Passenger Compartment Violations

Bulan

Xu²,

Loce³

et al. 2017

Computer Vision and Imaging in Intelligent Transportation Systems

View full text Add to dashboard Cite

Cross‐modal retrieval with dual multi‐angle self‐attention

Zheng

Zhang

et al. 2020

Asso for Info Science & Tech

View full text Add to dashboard Cite

In recent years, cross‐modal retrieval has been a popular research topic in both fields of computer vision and natural language processing. There is a huge semantic gap between different modalities on account of heterogeneous properties. How to establish the correlation among different modality data faces enormous challenges. In this work, we propose a novel end‐to‐end framework named Dual Multi‐Angle Self‐Attention (DMASA) for cross‐modal retrieval. Multiple self‐attention mechanisms are applied to extract fine‐grained features for both images and texts from different angles. We then integrate coarse‐grained and fine‐grained features into a multimodal embedding space, in which the similarity degrees between images and texts can be directly compared. Moreover, we propose a special multistage training strategy, in which the preceding stage can provide a good initial value for the succeeding stage and make our framework work better. Very promising experimental results over the state‐of‐the‐art methods can be achieved on three benchmark datasets of Flickr8k, Flickr30k, and MSCOCO.

show abstract

Fisher Kernels on Visual Vocabularies for Image Categorization

Cited by 1,376 publications

References 9 publications

Incremental kernel learning for active image retrieval without global dictionaries

Incremental kernel learning for active image retrieval without global dictionaries

Detection of Passenger Compartment Violations

Cross‐modal retrieval with dual multi‐angle self‐attention

Contact Info

Product

Resources

About