ECE Team Wins First Place at ACM Multimedia International Conference
A Northeastern team was awarded first place at the 2016 Association for Computing Machinery (ACM) Multimedia Conference MSR Image Recognition Grand Challenge. Along with a certificate and trophy, the award comes with a $1,000 bonus.
Microsoft Research hosts a series of Image Recognition (Retrieval) Grand Challenges to develop image recognition systems based on real world large scale data. This year, Microsoft released MS-Celeb-1M, a large scale real world face image dataset to the public, encouraging researchers to develop the best face recognition techniques to recognize one million people.
In its V1.0 version, the dataset contains 10 million celebrity face images for the top 100 thousands celebrities. The contestants are asked to develop image recognition system based on, but not limited to, the datasets provided by the Challenge (as training data) to recognize 1 million celebrities from their face images. To handle the large scale face recognition problem, the Northeastern team proposes an independent softmax model based on deep convolutional neural network. The approach achieved the highest accuracy in 14 teams from all over the world.
The ECE Department is proud to congratulate Professor Yun Raymond Fu, PhD Candidate Yue Wu, Postdoc Jun Li, and Postdoc Yu Kong.