Muhammad Huzaifa
Muhammad Huzaifa
AmbisonicEncoder's process() function only encodes a mono-signal. If we wanted to encode several different sound sources, how would we go about doing that? Thanks!
Even though our VIO setup is accurate over longer time frames (as confirmed by trajectory plots), it suffers from jitter over shorter durations. We should add a low pass filter...
Closes #149. Draft PR contingent upon response to [this question](https://github.com/rpng/open_vins/issues/247). The response may also motivate a new integrator.
The OpenGL code inside ILLIXR can be considerably cleaned up. TODOs: - [x] Remove unnecessary GLFW dependencies from gldemo and timewarp (it seems GLFW is only required for time) -...
In addition to what data needs to be passed from the SLAM to the integrator to the predictor, the _format_ of the data also needs to be standardized. This includes...
Our eye tracking code is currently written in Python. It needs to be ported to C++ for both performance and integration reasons. PyTorch has a nice C++ API that is...
All the SLAM dataset readers -- offline_imu_cam, ground_truth_slam, pose_lookup -- roll their own reader, and the readers all look the same. We should have a common `dataset_reader` class that everyone...
Rationale: full VI-SLAM system with loop closure. ORB-SLAM3 has maps that stay persistent across runs, and we will need an efficient mechanism in ILLIXR to store/retrieve them.
If user specifies both offline_imu_cam and ZED plugins, the system should fault. Currently, it might assign both feeds to the same switchboard topic, resulting in a spliced feed.