CVPR 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Congratulations to our InfleXion Lab for the paper titled “Texture and Shaped biased Two-Stream Networks for Clothing Classification and Attribute Recognition” was accepted for the main conference paper at the IEEE CVPR 2020 (Computer Vision and Pattern Recognition). The research topic comes from the specific application scenarios provided by a start-up company LeTote China.

In order to help the apparel industry to solve the problem of smart clothing matching, the researchers have explored an algorithm that more accurately and clearly identifies clothing attributes.  They explored the usefulness of landmarks and found that landmarks can assist in extracting shape features; and using landmarks for joint learning can increase classification and recognition accuracy effectively. They found that the texture features have an impelling effect on these tasks, and the pre-trained ImageNet model has good performance in extracting texture features.

Therefore, researchers propose to use two-streams to enhance the extraction of shapes and textures, respectively. In addition, this paper proposed a simple implementation, Texture and Shape biased Fashion Networks (TS-FashionNet). Comprehensive and rich experiments demonstrate the effectiveness of proposed model.


Texture and Shape Biased Two Stream Networks for Clothing Classification and Attribute Recongition

TS-FashionNet

CVPR is the premier annual computer vision event comprising the main conference and several co-located workshops and short courses. With its high quality and low cost, it provides an exceptional value for students, academics and industry researchers.

This year CVPR received a record number of 6656 submissions and accepted 1470 papers — a 22% acceptance rate — so the process was highly selective.

Despite increasing COVID-19 concerns, organizers of this year’s Conference on Computer Vision and Pattern Recognition (CVPR 2020) say the leading AI conference remains scheduled as planned, June 14 to 19 at the Washington State Convention Center in Seattle.

InfleXion Lab at Tsinghua Shenzhen International Graduate School

Powered by ParticleX, InfleXion Lab integrates the network of Tsinghua Shenzhen International Graduate School into the field of scientific research and the resources of Shine Works. InfleXion Lab promotes research collaboration between enterprises and universities, fosters the industrialization of scientific research, and introduces innovative technologies from around the world to the Greater Bay Area.