A data driven method for feature transformation

Dikmen, Mert; Hoiem, Derek; Huang, Thomas S.

doi:10.1109/cvpr.2012.6248069

Cited by 16 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these approaches do not learn the variable deformation properties of body parts. The approach in [7] learns features and a part-based model sequentially but not jointly.…”

Section: Related Workmentioning

confidence: 99%

Joint Deep Learning for Pedestrian Detection

Ouyang

Wang

2013

2013 IEEE International Conference on Computer Vision

591

359

View full text Add to dashboard Cite

show abstract

“…However, these approaches do not learn the variable deformation properties of body parts. The approach in [7] learns features and a part-based model sequentially but not jointly.…”

Section: Related Workmentioning

confidence: 99%

Joint Deep Learning for Pedestrian Detection

Ouyang

Wang

2013

2013 IEEE International Conference on Computer Vision

591

359

View full text Add to dashboard Cite

show abstract

“…The image classification performance they report significantly lags modern SIFT-based models such as those described in [6], despite the fact that they learn a multi-layered feature representation. The power of learned patch-level features has also been demonstrated recently in [5, 9, 24]. Using mini-epitomes instead of image patches could also prove beneficial in their setting.…”

Section: Related Workmentioning

confidence: 85%

Modeling Image Patches with a Generic Dictionary of Mini-epitomes

Papandreou

Chen

Yuille

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

The goal of this paper is to question the necessity of features like SIFT in categorical visual recognition tasks. As an alternative, we develop a generative model for the raw intensity of image patches and show that it can support image classification performance on par with optimized SIFT-based techniques in a bag-of-visual-words setting. Key ingredient of the proposed model is a compact dictionary of mini-epitomes, learned in an unsupervised fashion on a large collection of images. The use of epitomes allows us to explicitly account for photometric and position variability in image appearance. We show that this flexibility considerably increases the capacity of the dictionary to accurately approximate the appearance of image patches and support recognition tasks. For image classification, we develop histogram-based image encoding methods tailored to the epitomic representation, as well as an “epitomic footprint” encoding which is easy to visualize and highlights the generative nature of our model. We discuss in detail computational aspects and develop efficient algorithms to make the model scalable to large tasks. The proposed techniques are evaluated with experiments on the challenging PASCAL VOC 2007 image classification benchmark.

show abstract

“…(4) as a similarity measure is popularly used in the applications such as document retrieval [23], image decomposition [19], image quality assessment [20] and feature transformation [24]. Mathematically, dot product is also called "cosine similarity" and is naturally tied to the popular variant of the k-means algorithms known as spherical k-means [25].…”

Section: Gradient Correlation Similarity (Gcs) Measurementioning

confidence: 99%

GcsDecolor: Gradient Correlation Similarity for Efficient Contrast Preserving Decolorization

Liu

Xie

et al. 2015

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

This paper presents a novel gradient correlation similarity (Gcs) measure-based decolorization model for faithfully preserving the appearance of the original color image. Contrary to the conventional data-fidelity term consisting of gradient error-norm-based measures, the newly defined Gcs measure calculates the summation of the gradient correlation between each channel of the color image and the transformed grayscale image. Two efficient algorithms are developed to solve the proposed model. On one hand, due to the highly nonlinear nature of Gcs measure, a solver consisting of the augmented Lagrangian and alternating direction method is adopted to deal with its approximated linear parametric model. The presented algorithm exhibits excellent iterative convergence and attains superior performance. On the other hand, a discrete searching solver is proposed by determining the solution with the minimum function value from the linear parametric model-induced candidate images. The non-iterative solver has advantages in simplicity and speed with only several simple arithmetic operations, leading to real-time computational speed. In addition, it is very robust with respect to the parameter and candidates. Extensive experiments under a variety of test images and a comprehensive evaluation against existing state-of-the-art methods consistently demonstrate the potential of the proposed model and algorithms.

show abstract

A data driven method for feature transformation

Cited by 16 publications

References 24 publications

Joint Deep Learning for Pedestrian Detection

Joint Deep Learning for Pedestrian Detection

Modeling Image Patches with a Generic Dictionary of Mini-epitomes

GcsDecolor: Gradient Correlation Similarity for Efficient Contrast Preserving Decolorization

Contact Info

Product

Resources

About