RF Liang

Hi, I recently tested this model by using some hand-drawn sketches (rather sketches by sketch generators). I found the generated images look much worse than using generated sketches. For example...

DTU per-scene optimization

1

Hi @Xharlie, Thanks for your efforts in this project. I tried reproducing some of the results listed in your paper about the DTU dataset these days. I noticed that the...

Replace pycuda with torch extensions

6

Hi, I made a fork of this repo (https://github.com/nexuslrf/pointnerf/tree/master/models/neural_points/c_ext) and replaced all pycuda modules with torch extension, the CUDA kernels are almost the same as before. I verified the modifications...

Hi, I saw the encoder has an `ash` option in https://github.com/ashawkey/torch-ngp/blob/35a8ae54063e794b40a9f0e33df2b823856ffdf6/encoding.py#L71 However, the related implementation of `AshEncoder` is not available. I wonder whether this Ash encoder is based on this...

Camera Optimizer has different pose adjustments when applied to Raybundle and Camera.

4

Hi developers, I found a pretty tricky computation inconsistency issue on `camera_optimizer`'s `apply_to_*` functions: https://github.com/nerfstudio-project/nerfstudio/blob/babf57718c75e7d04f1b977f035db67c02a9c20a/nerfstudio/cameras/camera_optimizers.py#L148-L168 The rotation and translation shown in the above two functions are completely different. Assume the...