Loading...
Search for:
saliency-map
0.053 seconds
Class attention map distillation for efficient semantic segmentation
, Article 1st International Conference on Machine Vision and Image Processing, MVIP 2020, 19 February 2020 through 20 February 2020 ; Volume 2020-February , 2020 ; Kasaei, S ; Sharif University of Technology
IEEE Computer Society
2020
Abstract
In this paper, a novel method for capturing the information of a powerful and trained deep convolutional neural network and distilling it into a training smaller network is proposed. This is the first time that a saliency map method is employed to extract useful knowledge from a convolutional neural network for distillation. This method, despite of many others which work on final layers, can successfully extract suitable information for distillation from intermediate layers of a network by making class specific attention maps and then forcing the student network to mimic producing those attentions. This novel knowledge distillation training is implemented using state-of-the-art DeepLab and...
Design a Content-Based Color Image Retrieval Using Attention Driven Saliency Map
, M.Sc. Thesis Sharif University of Technology ; Fatemizadeh, Emadeddin (Supervisor)
Abstract
Content Based Image Retrieval (CBIR) is in fact an image search engine which Operates on image Context . in this thesis (project) the aim was to use the Visual attention of humans in detecting the objects in image. in this ability first a salient image of the most important things in the image would be created And after an initial separation , for the final recognition the other features (details) in the image will be used It’s a while that the use of Visual attention models and saliency maps in designing the interfaces between humans and machines has been considered widely. This fact in the design of CBIR systems has not a good background (satisfying history). In this thesis I have...
Class Attention Map Distillation In Semantic Segmentation
, M.Sc. Thesis Sharif University of Technology ; Kasaei, Shohreh (Supervisor)
Abstract
Semantic segmentation is the tash of labeling each pixel of an input image. It is one of the main problems in computer vision and plays an important role in scene understanding. State of the art methods of solving it are based on Convolutional Neural Networhs (CNNs). While many real world tasks like autodriving cars and robot navigation require fast and lightweight models, CNNs inherently tend to give beter accuracy when they are deeper and bigger, and this has raised interest in designing compact networks. Knowledge distillation is one of the popular methods of training compact networhs and helps to transfer a big and powerful network’s knowledge to a small and compact one. In this research...
Deep relative attributes
, Article 13th Asian Conference on Computer Vision, ACCV 2016, 20 November 2016 through 24 November 2016 ; Volume 10115 LNCS , 2017 , Pages 118-133 ; 03029743 (ISSN); 9783319541921 (ISBN) ; Noury, E ; Adeli, E ; Sharif University of Technology
Springer Verlag
2017
Abstract
Visual attributes are great means of describing images or scenes, in a way both humans and computers understand. In order to establish a correspondence between images and to be able to compare the strength of each property between images, relative attributes were introduced. However, since their introduction, hand-crafted and engineered features were used to learn increasingly complex models for the problem of relative attributes. This limits the applicability of those methods for more realistic cases. We introduce a deep neural network architecture for the task of relative attribute prediction. A convolutional neural network (ConvNet) is adopted to learn the features by including an...
A modified saliency detection for content-aware image resizing using cellular automata
, Article Proceedings of the 2010 International Conference on Signal and Image Processing, ICSIP 2010, 15 December 2010 through 17 December 2010 ; 2010 , Pages 175-179 ; 9781424485949 (ISBN) ; Jamzad, M ; Sharif University of Technology
Abstract
It is often required that image resizing be done brightly in order to preserve important content. Some image resizing techniques like scaling and cropping fail to identify and protect important objects, or they produce non-photorealistic images. But content aware image resizing schemes aim to change image aspect ratios while preserving visually outstanding features. In this paper, a novel method for content aware resizing is presented. Seam carving, an effective image resizing algorithm, fails to protect important objects in images, when either the energy content of the objects are low with respect to their surroundings, or, the number of seams removed are very large. Using saliency map as...
A framework for content-based human brain magnetic resonance images retrieval using saliency map
, Article Biomedical Engineering - Applications, Basis and Communications ; Volume 25, Issue 4 , 2013 ; 10162372 (ISSN) ; Fatemizadeh, E ; Badie, K ; Sharif University of Technology
2013
Abstract
Content-based image retrieval (CBIR) makes use of low-level image features, such as color, texture and shape, to index images with minimal human interaction. Considering the gap between low-level image features and the high-level semantic concepts in the CBIR, we proposed an image retrieval system for brain magnetic resonance images based on saliency map. The saliency map of an image contains important image regions which are visually more conspicuous by virtue of their contrast with respect to surrounding regions. First, the proposed approach exploits the ant colony optimization (ACO) technique to measure the image's saliency through ants' movements on the image. The textural features are...