gtsfm
gtsfm copied to clipboard
borglab
End-to-end SFM framework based on GTSAM
This PR creates a `gtsfm` command that users can utilize to run both the React JS frontend and the python backend. The command follows these steps: 1. Start the React...
First pass at using GTSfM's poses and camera parameters as input to Nerfstudio. These screenshots come from Nerfstudio's Viewer.
Here is a demo of the door reconstruction on Colab. I was able to visualize the final reconstruction with `visu3d` by using the source code instead of the pypi wheel....
We want to move all code from `MultiViewOptimizer` class that does not call `create_computation_graph` functions. This PR moves `prune_to_largest_connected_component` calculation into `ViewGraphEstimator`.
This PR moves the `init_cameras` function from `MultiViewOptimizer` into `DataAssociation`. The reason behind this is to move all code from `MultiViewOptimizer` which is not a call to `create_computation_graph` functions. This...
The Bundle Adjustment class is used twice in the pipeline, which forced us to define a dummy class for two-view BA and put the global BA UiMetadata in the actual...
The process graph introduced in #540 exposes some places where our code does not align with a pure bipartite graph model, where processes produce and consume whole products, without modification....
We should fix the naming convention for different modules/plates: both for functions as well as variables/argument. 1. Functions: 1.1. The modules should provide separate functions to run the algorithm to...
