Vision Wins Best Open Source Software Award at ACM Multimedia

Audio and Multimedia Director Gerald Friedland accepting the award on behalf of Vision Group members

^{December 9, 2014}
Members of the Vision Group won the Best Open Source Software Award at the ACM International Conference on Multimedia in November. They were recognized for their paper describing Caffe, or Convolutional Architecture for Fast Feature Embedding, a deep learning framework that can be used to build multimedia analysis tools. When used for image analysis, Caffe can classify an image in about two milliseconds and process more than 40 million images in a day. The framework has also been used by the Audio and Multimedia Group in an audio analysis tool, audioCaffe, which it is using to process data from the Yahoo Flickr Creative Commons 100 Million Dataset. Caffe is developed by the Berkeley Vision Learning Center, which comprises professors and researchers from UC Berkeley and ICSI.

For more information, visit the Caffe web page.

Related Paper: Caffe: Convolutional Architecture for Fast Feature Embedding. Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, and T. Darrell. Proceedings of the 22nd ACM International Conference on Multimedia (Multimedia 2014), Orlando, Florida, pp. 675-678. Winner of the ACM Multimedia Open Source Software Competition, November 2014