Bilateral grid slicing. Supports 2-D, 3-D, 4-D, and 5-D input. For the 2-D, 3-D, and 4-D cases, please refer to `slice`. For the 5-D c

Low-rank 4D bilateral grid slicing. Args: xyz (torch.Tensor): The xyz coordinates with shape $(..., 3)$. rgb (torch.T

Adjust camera pose based on deltas. Args: camtoworlds: (..., 4, 4) embed_ids: (...,) Returns: up

Adjust appearance based on embeddings. Args: features: (N, feature_dim) embed_ids: (C,) dirs: (C, N, 3)

Generate an elliptical render path based on the given poses.

Copied from Plenoxels Continuous learning rate decay function. Adapted from JaxNeRF The returned rate is lr_init when step=0 and lr_fina

Initialize and return the running state for this strategy. The returned state should be passed to the `step_pre_backward()` and `step

Initialize and return the running state for this strategy. The returned state should be passed to the `step_pre_backward()` and `step

Initialize and return the running state for this strategy. The returned state should be passed to the `step_pre_backward()` and `step

Load image sequence and convert to tensor sequence Args: file_pattern: Complete file path pattern, e.g. "results/mpeg151/video_a

Inverse transform the splats parameters to original domain Args: splats_list (List[Dict]): splats parameters of all frame

Inverse transform the splats parameters to original domain Args: splats_list (List[Dict]): splats parameters of all frame

Transform the splats parameters to new domain e.g. for SH coefficients, transform them from RGB domain to YCbCr domain for q

Transform the splats parameters to new domain e.g. for SH coefficients, transform them from RGB domain to YCbCr domain for q

Perspective projection on Gaussians. DEPRECATED: please use `proj` with `ortho=False` instead. Args: means: Gaussian means. [C, N, 3]

Quantize input splats and return quantized tensors with context.