Efficient Object Pixel-Level Categorization Using Bag of Features

Aldavert, David; Ramisa, Arnau; Toledo, Ricardo; Mántaras, Ramon López de

doi:10.1007/978-3-642-10331-5_5

Cited by 4 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Later, there was a research to integrate BoW technique into visual content retrieval under the designation of Video Google [2]. In addition BoW technique was applied for scene recognition or classification, and object segmentation [3][4][5][6][7][8][9][10]. The image is represented by histogram of visual words like document representation, where the visual word is a region descriptor extracted from local patches of an image.…”

Section: Introductionsupporting

confidence: 47%

Bag-of-features signature using invariant region descriptor for object retrieval

Cho

Yang

Jeong

et al. 2011

2011 17th Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV)

View full text Add to dashboard Cite

In recent years, a content-based method such as 'bagof-features' (BoF) is coming to the fore as an object recognition and classification technique. This paper proposes a BoF signature using invariant region descriptor for object retrieval. The region descriptors are extracted from dense sampled regions in the training images. These descriptors are quantized by hierarchical k-means clustering in a vocabulary tree of visual words. Each image is represented by occurrence of visual words, and then we use linear combination distance measure in the matching. In the experiments, we use object images that are taken in different condition to evaluate the retrieval performance. The test results show that the proposed method outperforms the BoF method using SURF descriptor. The proposed method searches 2.9 correct images among 3 on average up to the top 3% rank in database. Therefore, the proposed method is considered as an effective technique in terms of retrieval accuracy.

show abstract

Section: Introductionsupporting

confidence: 47%

Bag-of-features signature using invariant region descriptor for object retrieval

Cho

Yang

Jeong

et al. 2011

2011 17th Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV)

View full text Add to dashboard Cite

show abstract

“…A HOG features describes the narrow shape of objects, having edge information at different cells. In plain reigns, the histogram of oriented gradients [9] will have flatter dimensions for example: ground or building wall whereas in the borders, most one of the elements in the histogram has a largest value and it indicates edge direction. Even though the images are normalize to the place and scale, the positing of main features will not be registered with the same position in the grid.…”

Section: Figure 1: Shows the Block Diagram Of The Proposed Methodsmentioning

confidence: 99%

“…Aldavert et al [9] proposed object categorization method using bag of features and SVM. Hierarchical K-Means and the Extremely Randomized Forest are used to measure the performance.…”

Section: Introductionmentioning

confidence: 99%

Object Classification using SVM and KD-Tree

N¹,

Murali²

2020

IJRTE

View full text Add to dashboard Cite

In this proposed work, we presented a system to classify the object. Firstly, the given images are segmented using Region merging Segmentation method. Later the background eliminated images are divided into number of blocks viz., 4, 16, 32. The features like Scale Invariant Feature Transform (SIFT) and Histogram of Gradients (HOG) are extracted from divided blocks of size 4, 16, 32. To measure the strength of proposed method we compare the Classification vs Retrieval using Support Vector Machine and KD Tree. We conducted the experimentation on Caltech 101 data set. To study the effect of accuracy in classification we pick images from database randomly. The Performance revels that the SVM achieves good performance.

show abstract