Traditional Visual Words Research Articles

Spatiotemporal dynamics of orthographic and lexical processing in the ventral visual pathway.

Reading is a rapid, distributed process that engages multiple components of the ventral visual stream. To understand the neural constituents and their interactions that allow us to identify written words, we performed direct intra-cranial recordings in a large cohort of humans. This allowed us to isolate the spatiotemporal dynamics of visual word recognition across the entire left ventral occipitotemporal cortex. We found that mid-fusiform cortex is the first brain region sensitive to lexicality, preceding the traditional visual word form area. The magnitude and duration of its activation are driven by the statistics of natural language. Information regarding lexicality and word frequency propagates posteriorly from this region to visual word form regions and to earlier visual cortex, which, while active earlier, show sensitivity to words later. Further, direct electrical stimulation of this region results in reading arrest, further illustrating its crucial role in reading. This unique sensitivity of mid-fusiform cortex to sub-lexical and lexical characteristics points to its central role as the orthographic lexicon-the long-term memory representations of visual word forms.

Multi-order visual phrase for scalable partial-duplicate visual search

Visual phrase considers multiple visual words and captures extra spatial clues among them. Thus, visual phrase shows better discriminative power than single visual word in image retrieval and matching. Not withstanding their success, existing visual phrases still show obvious shortcomings: (1) limited flexibility, i.e., visual phrases are considered for matching only if they contain the same number of visual words; (2) large quantization error and low repeatability, i.e., quantization errors in visual words are aggregated in visual word combinations and visual phrases, making them harder to be matched than single visual words. To avoid these issues, we propose multi-order visual phrase (MVP) which contains two complementary clues: center visual word quantized from the local descriptor of each image keypoint and the visual and spatial clues of multiple nearby keypoints. Two MVPs are flexibly matched by first matching their center visual words, then estimating a match confidence by checking the spatial and visual consistency of their neighbor keypoints. Therefore, center visual word matching equals to traditional visual word matching, but the neighbor spatial and visual clues checking significantly boosts the discriminative power. MVP does not scarify the repeatability of single visual word and is more robust to quantization error than existing visual phrases. We test our approach in three image retrieval tasks on UKbench, Oxford5K, and 1 million distractor images collected from Flickr. Comparisons with recent retrieval approaches and existing visual phrase features clearly demonstrate the competitive accuracy and significantly better efficiency of MVP.

Scene Classification Algorithm Based on Contextual Semantic Information Extended PLSA Model

In this paper, a novel scene classification algorithm based on contextual visual words combined with extended PLSA model is proposed in which the traditional visual words by contextual information are improved. First, the EICS-LBP (Edge Improved Center Symmetric Local Binary Pattern) and statistical domain color features are extracted from the edge dense sampling region and its 8 neighborhood followed by the computed normalized histogram intersection and the creation of the new feature by linear combination of local information and its contextual intersection information. The visual vocabulary is formed by clustering from dense sampling regions respectively. Then the Bag-Of-Words model is used to describe the image which gives us richer information of scene images and reduces ambiguities and errors. Finally, the potential semantics by extended PLSA model and the confusion table by KNN classifier are obtained. Experiment results show that this method could achieve a higher accuracy, especially performing well in multi-edge color images with contextual information. Besides, it does not require manual annotation on the scene.

Region Contextual Visual Words for scene categorization

This paper proposes a method for scene categorization by integrating region contextual information into the popular Bag-of-Visual-Words approach. The Bag-of-Visual-Words approach describes an image as a bag of discrete visual words, where the frequency distributions of these words are used for image categorization. However, the traditional visual words suffer from the problem when faced these patches with similar appearances but distinct semantic concepts. The drawback stems from the independently construction each visual word. This paper introduces Region-Conditional Random Fields model to learn each visual word depending on the rest of the visual words in the same region. Comparison with the traditional Conditional Random Fields model, there are two areas of novelty. First, the initial label of each patch is automatically defined based on its visual feature rather than manually labeling with semantic labels. Furthermore, the novel potential function is built under the region contextual constraint. The experimental results on the three well-known datasets show that Region Contextual Visual Words indeed improves categorization performance compared to traditional visual words.

Scene categorization with multiscale category-specific visual words

We propose a novel scene categorization method based on multiscale category-specific visual words. The novelty of the proposed method lies in two aspects: (1) visual words are quantized in a multiscale manner that combines the global-feature-based and local-feature-based scene categorization approaches into a uniform framework; (2) unlike traditional visual word creation methods, which quantize visual words from the entire set of training, we form visual words from the training images grouped in different categories and then collate visual words from different categories to form the final codebook. This generation strategy is capable of enhancing the discriminative ability of the visual words, which is useful for achieving better classification performance. The proposed method is evaluated over two scene classification data sets with 8 and 13 scene categories, respectively. The experimental results show that the classification performance is significantly improved by using the multiscale category-specific visual words over that achieved by using the traditional visual words. Moreover, the proposed method is comparable with the best methods reported in previous literature in terms of classification accuracy rate (88.81% and 85.05% accuracy rates for data sets 1 and 2, respectively) and has the advantage in simplicity.