arXiv subjects

Xiaofu Jin

Publications and source records attributed to Xiaofu Jin.

20 records · Page 2Linked to original sources

arXiv2022

"Merging Results Is No Easy Task": An International Survey Study of Collaborative Data Analysis Practices Among UX Practitioners

Analysis is a key part of usability testing where UX practitioners seek to identify usability problems and generate redesign suggestions. Although previous research reported how analysis was conducted, the findings were typically focused on individual analysis or based on a small number of professionals in specific geographic regions. We conducted an online international survey of 279 UX practitioners on their practices and challenges while collaborating during data analysis. We found that UX practitioners were often under time pressure to conduct analysis and adopted three modes of collaboration: independently analyze different portions of the data and then collaborate, collaboratively analyze the session with little or no independent analysis, and independently analyze the same set of data and then collaborate. Moreover, most encountered challenges related to lack of resources, disagreements with colleagues regarding usability problems, and difficulty merging analysis from multiple practitioners. We discuss design implications to better support collaborative data analysis.

cs.HC↗

arXiv2019

An End-to-End Neural Network for Image Cropping by Learning Composition from Aesthetic Photos

As one of the fundamental techniques for image editing, image cropping discards unrelevant contents and remains the pleasing portions of the image to enhance the overall composition and achieve better visual/aesthetic perception. In this paper, we primarily focus on improving the accuracy of automatic image cropping, and on further exploring its potential in public datasets with high efficiency. From this respect, we propose a deep learning based framework to learn the objects composition from photos with high aesthetic qualities, where an anchor region is detected through a convolutional neural network (CNN) with the Gaussian kernel to maintain the interested objects' integrity. This initial detected anchor area is then fed into a light weighted regression network to obtain the final cropping result. Unlike the conventional methods that multiple candidates are proposed and evaluated iteratively, only a single anchor region is produced in our model, which is mapped to the final output directly. Thus, low computational resources are required for the proposed approach. The experimental results on the public datasets show that both cropping accuracy and efficiency achieve the state-ofthe-art performances.

cs.CV↗