Exploiting color cues to improve person re-identification

Author(s):  
C. Chahla ◽  
H. Snoussi ◽  
F. Abdallah ◽  
F. Dornaika
Keyword(s):  
2021 ◽  
pp. 1-29
Author(s):  
Cameron Brick ◽  
Alexandra L.J. Freeman

Abstract Policy decisions have vast consequences, but there is little empirical research on how best to communicate underlying evidence to decision-makers. Groups in diverse fields (e.g., education, medicine, crime) use brief, graphical displays to list policy options, expected outcomes and evidence quality in order to make such evidence easy to assess. However, the understanding of these representations is rarely studied. We surveyed experts and non-experts on what information they wanted and tested their objective comprehension of commonly used graphics. A total of 252 UK residents from Prolific and 452 UK What Works Centre users interpreted the meaning of graphics shown without labels. Comprehension was low (often below 50%). The best-performing graphics combined unambiguous metaphorical shapes with color cues and indications of quantity. The participants also reported what types of evidence they wanted and in what detail (e.g., subgroups, different outcomes). Users particularly wanted to see intervention effectiveness and quality, and policymakers also wanted to know the financial costs and negative consequences. Comprehension and preferences were remarkably consistent between the two samples. Groups communicating evidence about policy options can use these results to design summaries, toolkits and reports for expert and non-expert audiences.


2013 ◽  
Vol 23 (3) ◽  
pp. 341-348 ◽  
Author(s):  
Tine De Bock ◽  
Mario Pandelaere ◽  
Patrick Van Kenhove
Keyword(s):  

1979 ◽  
Vol 48 (2) ◽  
pp. 479-488 ◽  
Author(s):  
Robert M. Boynton ◽  
Stanley Dolensky

In a search and recognition task utilizing real-world objects, the usefulness of color cues is observed to depend upon the availability of alphanumeric information. Following a 45-sec. inspection of a randomly selected collection of 17 books, spread on a table with their titles exposed, subjects were asked to identify as many of these as possible during a test period beginning 3 min. later in which 17 decoys were also present. Some subjects wore glasses with red filters during the test and inspection periods. The color blindness thereby introduced did not impair their performance in comparison with control subjects who were able to utilize normal color vision. Moreover, the introduction of color during the test period impaired the performance of subjects who had been deprived of color cues during inspection. It was concluded that subjects paid attention mostly to book titles and for that reason did not use other cues, including color, to much advantage. In a second experiment, where titles were obscured, subjects with normal color vision performed much better than those who were made color blind during the inspection or test periods. The results of both experiments are generally consistent with predictions based on experiments which have used abstract stimulus materials.


2012 ◽  
pp. 387-397 ◽  
Author(s):  
Qingyuan Wang ◽  
Junbiao Pang ◽  
Lei Qin ◽  
Shuqiang Jiang ◽  
Qingming Huang
Keyword(s):  

2018 ◽  
Vol 122 ◽  
pp. 23-31 ◽  
Author(s):  
Alexander Skulmowski ◽  
Günter Daniel Rey
Keyword(s):  

2014 ◽  
Vol 27 (3-4) ◽  
pp. 207-223 ◽  
Author(s):  
Devin Z. Shermer ◽  
Carmel A. Levitan

Color cues can influence the experience of flavor, both by influencing identification and perceived intensity of foods. Previous research has largely focused on the crossmodal influence of vision upon taste or olfactory cues. It is plausible that color cues could also affect perceived trigeminal sensation; these studies demonstrate a crossmodal influence of color on piquancy. In our first two experiments, participants rated the spiciness of images of salsas that were adjusted to vary in color and intensity. We found that red was associated with significantly higher ratings of expected spice than blue, and that darker reds were expected to be spicier than lighter reds. In our third experiment, participants tasted and then rated the spiciness of each of four salsas (with two levels of color and of piquancy) when sighted and when blindfolded. Spiciness ratings were unaffected by differing colors when the salsa was mild, but when the piquancy was increased, a lack of increase in color corresponded to a depressed spiciness. These results can be explained using a model of assimilation and contrast. Taken together, our findings show that in our US sample, there is a crossmodal correspondence between visual and trigeminal senses that can influence perception of spiciness.


Sign in / Sign up

Export Citation Format

Share Document