Spatial Order Information Research Articles

On Exploring Multiplicity of Primitives and Attributes for Texture Recognition in the Wild.

Texture recognition is a challenging visual task since its multiple primitives or attributes can be perceived from the texture image under different spatial contexts. Existing approaches predominantly built upon CNN incorporate rich local descriptors with orderless aggregation to capture invariance to the spatial layout. However, these methods ignore the inherent structure relation organized by primitives and the semantic concept described by attributes, which are critical cues for texture representation. In this paper, we propose a novel Multiple Primitives and Attributes Perception network (MPAP) that extracts features by modeling the relation of bottom-up structure and top-down attribute in a multi-branch unified framework. A bottom-up process is first proposed to capture the inherent relation of various primitive structures by leveraging structure dependency and spatial order information. Then, a top-down process is introduced to model the latent relation of multiple attributes by transferring attribute-related features between adjacent branches. Moreover, an augmentation module is devised to bridge the gap between high-level attributes and low-level structure features. MPAP can learn representation through jointing bottom-up and top-down processes in a mutually reinforced manner. Experimental results on six challenging texture datasets demonstrate the superiority of MPAP over state-of-the-art methods in terms of accuracy, robustness, and efficiency.

Time, space, and memory for order

Information about the order of items in a sequence can be conveyed either spatially or temporally. In the present investigation, we examined whether these different modes of presentation map onto compatible mental representations of serial order. We examined this issue in three immediate serial-recall experiments, in which participants recalled lists of letters in the temporal order in which they had appeared. Each letter in a to-be-remembered sequence was presented in a unique spatial position, with the order of these spatial positions progressing from either left to right or right to left. In this way, the visually presented lists contained both temporal and spatial order information. Recall of the temporal order information was more accurate with congruent spatial order information-that is, when the letters progressed from left to right, following the typical reading direction of English-than when the spatial order information was incongruent. These results suggest compatible representations of serial order when sequences are conveyed spatially and temporally.

Segmentation under Occlusions Using Selective Shape Prior

In this work, we address the problem of segmenting multiple objects, under possible occlusions, in a level set framework. A variational energy that incorporates a piecewise constant representation of the image in terms of the object regions and the object spatial order is proposed. To resolve occluded boundaries, prior knowledge of the shape of objects is also introduced within the segmentation energy. By minimizing the above energy, we solve the segmentation with depth problem, i.e., estimating the object boundaries, the object intensities, and the spatial order. The segmentation with depth problem was originally dealt with by the Nitzberg–Mumford–Shiota (NMS) variational formulation, which proposes segmentation energies for each spatial order. We discuss the relationships and show the computational advantages of our formulation over the NMS model, mainly due to our treatment of spatial order estimation within a single energy. A novelty here is that the spatial order information available in the image model is used to dynamically impose prior shape constraints only to occluded boundaries. Also presented are experiments on synthetic and real images that have promising results.

Hybrid order statistic filter and its application to image restoration

We introduce a new nonlinear filter for signal and image restoration, the hybrid order statistic (HOS) filter. Because it exploits both rank- and spatial-order information, the HOS realizes the advantages of nonlinear filters in edge preservation and reduction of impulsive noise components while retaining the ability of the linear filter to suppress Gaussian noise. We show that the HOS filter exhibits improved performance over both the linear Wiener and the nonlinear L filters in reducing mean-squared error in the presence of contaminated Gaussian noise. In many cases it also performs favorably compared with the Ll and rank-conditioned rank selection filters.

Spatial and temporal uncertainty in long-term memory

Two experiments examined the long-term retention of spatial order information. Both experiments produced substantial evidence of spatial memory, even though the retention test was unexpected, and the patterns of performance mimicked those found in the delayed retention of temporal order. The reconstruction of spatial and temporal position produced bow-shaped serial position curves and error generalization gradients that were roughly symmetrical around the true serial positions. These results, among others, seem to identify general characteristics of the mnemonic processing of dimensions, regardless of the time scales involved (e.g., immediate versus delayed).

Recall of order information by deaf signers: Phonetic coding in temporal order recall

To examine the claim that phonetic coding plays a special role in temporal order recall, deaf and hearing college students were tested on their recall of temporal and spatial order information at two delay intervals. The deaf subjects were all native signers of American Sign Language. The results indicated that both the deaf and hearing subjects used phonetic coding in short-term temporal recall, and visual coding in spatial recall. There was no evidence of manual or visual coding among either the hearing or the deaf subjects in the temporal order recall task. The use of phonetic coding for temporal recall is consistent with the hypothesis that recall of temporal order information is facilitated by a phonetic code.

Why is word recognition impaired by disorientation while the identification of single letters is not?

Past research has shown that speed of identifying single letters or digits is largely indifferent to orientation, whereas the recognition of single words or connected text is markedly disrupted by disorientation. In a series of four experiments, we attempted to reconcile these findings. The results suggest that disorientation does not impair the identification of the characters but disrupts the perception of their spatial arrangement. When spatial order information is critical for distinguishing between different stimuli, disorientation is disruptive because some rectification process is required to restore order information. Utilizing the similarity between the letter B and the number 13, we found strong effects of orientation when a stimulus was interpreted as the two-digit number 13 but not when interpreted as the single letter B. This, however, occurred only when the set of numbers to be classified included permutations of the same digits (Experiments 1 and 2). Odd-even decisions on single-digit and two-digit numbers (Experiment 3) yielded strong effects of stimulus orientation for order-dependent numbers (e.g., 32), weaker effects for order-independent numbers (e.g., 24), and none for repeated-digit (e.g., 22) or single-digit numbers. Classification time for two-letter Hebrew words evidenced strong effects of orientation for words that differed only in letter order but much weaker effects for words that had no letters in common, even when these were embedded within some words that did (Experiment 4).

Phonetic coding and order memory in relation to reading proficiency: A comparison of short-term memory for temporal and spatial order information

ABSTRACTSince children with reading disability are known to have problems using a phonetic memory strategy, it was expected that their recall of order would be inferior to that of good readers in situations where a phonetic strategy is optimal, i.e., when temporal order recall, but not necessarily spatial order recall, is required. On separate tests for retention of temporal sequence and spatial location, the good readers were better than the poor readers on the temporal order task as expected, but contrary to expectation, they maintained their superiority on the spatial task as well. Nevertheless, differences in the error patterns of the good and the poor readers are supportive of earlier evidence that links poor readers' short-term memory deficiencies to reduced effectiveness of phonetic representation.

Young children's use of temporal and spatial order information in short-term memory

Encoding and retrieval of temporal and spatial order information was investigated in the free recall and probed recognition performance of 3-, 4-, 5-, and 6-year-olds. In both tasks, three line drawings of familiar objects were presented successively in different spatial locations, so that temporal order was not confounded with spatial order. Subjects of all ages revealed an organization for the stimulus display based on the temporal sequence of presentation. With increasing age, subjects began recall more often with the first item so as to take advantage of this sequential organization. In the probed recognition task, subjects were asked questions requiring the use of temporal or spatial order information. Performance by the older subjects indicated that information about temporal and spatial order could be simultaneously encoded. The results are discussed in terms of a context theory derived from Estes's hierarchical association model.

Pattern coding of spatial order information in short-term memory

Three experiments were conducted to extend the finding that, whereas phonemic coding is employed for short-term temporal order recall, coding of temporal-spatial patterns is employed for short-term spatial order recall. In Experiment 1 subjects employed pattern coding for spatial order information even though item information had to be retained as well as spatial order information. In Experiment 2 temporal and spatial order recall were differentially affected by various interpolated tasks; temporal order recall was disrupted by tasks demanding vocalization, whereas spatial order recall was disrupted by tasks demanding processing of spatial information. In Experiment 3 a change in the method of response led subjects to employ pattern coding for temporal order recall without eliminating phonemic coding.

Coding of temporal-spatial patterns in short-term memory

Short-term retention of temporal and spatial order information were compared using the distractor paradigm. Whereas subjects employed phonemic coding when they recalled the temporal order of letters, they coded information about temporal-spatial patterns for spatial order recall. When phonemic coding was eliminated by forcing the subjects to pronounce irrelevant items during the presentation of the letter string, subjects employed the pattern strategy even in temporal order recall. Distractor tasks involving digit shadowing were effective at disrupting temporal order recall, but caused little disruption of spatial order recall. However, distractor tasks involving the processing of spatial order information did interfere with spatial order recall.

Short-term retention of temporal and spatial order

Retention of temporal and spatial order information were compared in a short-term recall situation where only order information was to be learned. Concave-upward bow-shaped serial position functions were found for temporal but not spatial positions, whether subjects recalled the temporal or the spatial order of the items. Retention of temporal order was superior to retention of spatial order. However, spatial and temporal order recall should be further compared with a design which eliminates an asymmetry present in this and previous studies.