search

NVlabs / instant-ngp

Instant neural graphics primitives: lightning fast NeRF and more
https://nvlabs.github.io/instant-ngp
Other
15.57k stars 1.89k forks source link

Realtime SLAM Capture #1053

Open keli95566 opened 1 year ago

keli95566 commented 1 year ago

Hi there! I noticed an amazing demo presented at SIGGRAPH real-time live this year: https://www.youtube.com/watch?v=MAXJWEoKbxY

Is the real-time SLAM capture + NERF training && rendering included in this repo as well?

Tom94 commented 1 year ago

Hi, no, the real-time SLAM pipeline is not released yet, although you can hook your own via the Python bindings, which already permit a dynamically growing dataset like in the demo.

keli95566 commented 1 year ago

Awesome! Thank you for the hint!

SSground commented 1 year ago

Hi, no, the real-time SLAM pipeline is not released yet, although you can hook your own via the Python bindings, which already permit a dynamically growing dataset like in the demo. hi,sir Can you give me a hint how to use the growing dataset?

TimSousa1 commented 1 year ago

Is there still any active development on it / will it ever be merged to the main branch?

Konstantysz commented 3 months ago

Hi, no, the real-time SLAM pipeline is not released yet, although you can hook your own via the Python bindings, which already permit a dynamically growing dataset like in the demo.

@Tom94 Hi, would like to provide some tip where to look for handling a dynamicly growing dataset in Python bindings? Is it related to function: 

void Testbed::override_sdf_training_data(pybind11::array_t<float> points, pybind11::array_t<float> distances);