Convolutional Architecture for Fast Feature Embedding (CAFFE) provides multimedia scientists and practitioners with a framework for state-of-the-art deep learning algorithms and a collection of reference models. It gives the best performance on object detection, evaluated on the hardest academic datasets: the PASCAL VOC 2007-2012 and the ImageNet 2013 Detection challenge. In this project, the student is required to realize the CAFFE architecture on FPGA in order to obtain a significant speed-up with reasonable object detection performance.
Research on the applications of CAFFE.
Research on the implementation of the CAFFE architecture based on the logic complexity
Research on the implementation of the CAFFE architecture based on the speed and performance.
Students who are interested in this project should be familiar with “HDL Programming Language” and have some background knowledge in “Logic Design” and “Convolutional Neural Network". Students must consult the supervisor before bidding on the project.
Lab allocations have not been finalised