Open Detection is a standalone open source project for object detection and recognition in images and 3d point clouds. In this proposal, I first make a short report of the state-of-the-art detection/recognition techniques. To improve the framework for CNNs for 2D detection, I mainly focus on two key tasks that remain to be addressed. The first one is to implement OD interface for training and testing neural network architectures with CNN and RNN. The second is to add several recognition/detection algorithms to OpenDetection. To this end, I decompose each task into several small tasks and propose my solutions respectively, aiming to perfect the project in an elegant way. I also create milestones for each small tasks. I can work on this project full-time during this summer and I’ve equipped myself with prior knowledge and programing experiences in object detection and deep learning, which makes me competent to this project.