Haiyang Mei

Title: Innovations by Haiyang Mei in Camouflaged Object Segmentation

Introduction

Haiyang Mei is a notable inventor based in Liaoning, China. He has made significant contributions to the field of computer vision, particularly in the area of image segmentation. With a total of 3 patents, his work focuses on advanced methods for segmenting camouflaged objects and mirrors in images.

Latest Patents

One of his latest patents is the "Camouflaged object segmentation method with distraction mining." This method introduces a novel approach to segmenting camouflaged object images by utilizing distraction mining. The process involves a multi-layer feature extractor, a positioning module, and a focusing module. The multi-layer feature extractor employs a traditional feature extraction network to gather various levels of contextual features. The positioning module uses RGB feature information to initially locate the camouflaged object, while the focusing module refines this information by eliminating distractions based on RGB features and preliminary position data. This innovative method enhances the segmentation of camouflaged objects by integrating distraction information into the process.

Another significant patent is the "Depth-aware method for mirror segmentation." This invention represents a breakthrough in scene segmentation within computer vision. The method, known as PDNet, includes a multi-layer feature extractor, a positioning module, and a delineating module. The multi-layer feature extractor captures contextual features, while the positioning module combines RGB and depth feature information to identify the mirror's position in the image. The delineating module then adjusts the boundary of the mirror using both RGB and depth information. This method is the first to successfully utilize both types of images for mirror segmentation, achieving excellent results even in complex environments.

Career Highlights

Haiyang Mei is affiliated with Dalian University of Technology, where he continues to advance his research in computer vision and image processing. His work has garnered attention for its innovative approaches and practical applications in various fields.

Collaborations

Some of his notable coworkers include Xin Yang and Xiaopeng Wei, who contribute to the collaborative efforts in research and development at Dalian University of Technology.

Conclusion

Haiyang Mei's contributions to the field of computer vision, particularly through his innovative patents, demonstrate his expertise and commitment to advancing technology in image segmentation. His work not only enhances the understanding of camouflaged objects and mirrors but also paves the way for future advancements in the field.