It is our immense pleasure to announce the beginning of a new PCL Code Sprint sponsored by NVidia: PCL-NVCS!
PCL Code Sprints are intended to rapidly advance the capabilities of the Point Cloud Library in a certain area/subject by offering stipends to talented student developers and pairing them with knowledgeable mentors for several months of accelerated software development. Projects will run for an initial period of 3 months, and all of the code produced will be BSD-licensed open source.
For this spring's PCL-NVidia Code Sprint, we have identified the following important areas for further development in PCL, and we are therefore searching for outstanding candidates (and mentors) to work on the following projects:
Human Body tracking: The high-level purpose of this project is to implement and analyze various techniques for extracting humans body poses in dense 3D data at high speed with CUDA. The applications are numerous, from gaming to gesture interfacing. The code sprint component will be a focused implementation that produces fast and reusable code. Some possible code sprint project specifications include:
- Given an estimate of the position of a person's head in an initial RGB-D frame, find and track the arms and head.
- Given a noisy estimate of upper body part labels in RGB-D, fit a cylinder-based upper body model to the depth data.
- Given an initial guess of a human skeleton pose and a RGB-D frame with noisy body part labels, try to refine the skeleton estimate. Then, given subsequent RGB-D frames and noisy part labels, continue to track the skeleton.
- Real-time visualization of large 3D point cloud data on mobile platforms: The purpose of this project is to couple data compression and streaming with 3D visualization for mobile platforms. Additional sensors available on mobile platforms such as IMUs will be integrated for the purpose of enhancing the visualization (i.e., keep track of the position of the tablet with respect to the virtual world and use physical gestures to zoom in, rotate, etc by moving the tablet and reading IMU data).
- Deformable object modeling: The purpose of this project is to create deformable models, such as human heads, or deformable toys using non-rigid registration on GPUs from dense 3D data. Applications include morphing and animation of non-rigid object models in real time. Computer graphics experience is a plus.
- Tegra3 optimizations for PCL and Kinect: This project will assemble a list of core components that form the basis of all modern 3D applications with PCL that need to be optimized for Tegra3/Android platforms using shaders, etc. Start with data acquisition (Kinect connected to a tablet), and go slowly towards Kinect Fusion on tablets. Knowledge of ARM optimizations is a plus.
The above projects will run for a period of 3 months divided into two terms. Each student developer will receive a stipend of $5000 per 3 months for their contributions, subject to satisfactory mid-term and final evaluations.
Potential candidates should submit the following information to email@example.com:
- a brief resume
- a list of existing PCL contributions (if any)
- a list of projects (emphasis on open source projects please) that they contributed to in the past
Advanced C++ programming skills are required for all projects. Please remember that this is a unique opportunity to work with some of the world's best 3D perception researchers!
Interested mentors should send a brief e-mail to the address mentioned above pointing out their time commitments and expertise in the field.