Files
nerfstudio/nerfactory/models/instant_ngp.py
T
Matthew Tancik a79ef3fb11 Breaking up nerfacc (#462)
* Breaking up nerfacc

* add background color
2022-09-17 17:33:54 -07:00

262 lines
9.8 KiB
Python

# Copyright 2022 The Plenoptix Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Implementation of Instant NGP.
"""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Dict, List, Literal, Optional, Tuple, Type
import nerfacc # pylint: disable=import-error
import torch
from torch.nn import Parameter
from torchmetrics import PeakSignalNoiseRatio
from torchmetrics.functional import structural_similarity_index_measure
from torchmetrics.image.lpip import LearnedPerceptualImagePatchSimilarity
from torchtyping import TensorType
from nerfactory.cameras.rays import RayBundle
from nerfactory.configs import base as cfg
from nerfactory.fields.instant_ngp_field import field_implementation_to_class
from nerfactory.fields.modules.field_heads import FieldHeadNames
from nerfactory.models.base import Model
from nerfactory.models.modules.ray_sampler import VolumetricSampler
from nerfactory.optimizers.loss import MSELoss
from nerfactory.utils import visualization
from nerfactory.utils.callbacks import (
TrainingCallback,
TrainingCallbackAttributes,
TrainingCallbackLocation,
)
@dataclass
class InstantNGPModelConfig(cfg.ModelConfig):
"""Instant NGP Model Config"""
_target: Type = field(
default_factory=lambda: NGPModel
) # We can't write `NGPModel` directly, because `NGPModel` doesn't exist yet
"""target class to instantiate"""
enable_collider: bool = False
"""Whether to create a scene collider to filter rays."""
collider_params: Optional[Dict[str, float]] = None
"""Instant NGP doesn't use a collider."""
field_implementation: Literal["torch", "tcnn"] = "tcnn" # torch, tcnn, ...
"""one of "torch" or "tcnn", or other fields in 'field_implementation_to_class'"""
# TODO(ethan): remove the density field specified here
enable_density_field: bool = False
"""Whether to create a density field to filter samples."""
num_samples: int = 1024
"""Number of samples in field evaluation."""
grid_resolution: int = 128
"""Resolution of the grid used for the field."""
cone_angle: float = 0.0
"""Should be set to 0.0 for blender scenes but 1./256 for real scenes."""
near_plane: float = 0.05
"""How far along ray to start sampling."""
randomize_background: bool = False
"""Whether to randomize the background color."""
class NGPModel(Model):
"""Instant NGP model
Args:
config: instant NGP configuration to instantiate model
"""
config: InstantNGPModelConfig
def __init__(self, config: InstantNGPModelConfig, **kwargs) -> None:
assert config.field_implementation in field_implementation_to_class
self.field = None
super().__init__(config=config, **kwargs)
def populate_modules(self):
"""Set the fields and modules."""
super().populate_modules()
# torch or tiny-cuda-nn version
self.field = field_implementation_to_class[self.config.field_implementation](self.scene_bounds.aabb)
# to match Ruilong's code naming
self.scene_aabb = Parameter(self.scene_bounds.aabb.flatten(), requires_grad=False)
# Sampler
self.sampler = VolumetricSampler(
aabb=self.scene_aabb,
density_fn=self.field.density_fn,
grid_resolution=self.config.grid_resolution,
num_samples=self.config.num_samples,
)
# losses
self.rgb_loss = MSELoss()
# metrics
self.psnr = PeakSignalNoiseRatio(data_range=1.0)
self.ssim = structural_similarity_index_measure
self.lpips = LearnedPerceptualImagePatchSimilarity()
def get_training_callbacks(
self, training_callback_attributes: TrainingCallbackAttributes
) -> List[TrainingCallback]:
return [
TrainingCallback(
where_to_run=[TrainingCallbackLocation.BEFORE_TRAIN_ITERATION],
update_every_num_iters=1,
func=self.sampler.occ_field.every_n_step, # will take in step
),
]
def get_param_groups(self) -> Dict[str, List[Parameter]]:
param_groups = {}
if self.field is None:
raise ValueError("populate_fields() must be called before get_param_groups")
param_groups["fields"] = list(self.field.parameters())
return param_groups
def get_outputs(self, ray_bundle: RayBundle):
assert self.field is not None
def query_fn(positions: TensorType["bs", 3], directions: TensorType["bs", 3], only_density=False):
assert self.field is not None
if only_density:
return self.field.density_fn(positions)
field_outputs = self.field.get_outputs_from_positions_and_direction(positions, directions)
rgbs = field_outputs[FieldHeadNames.RGB]
sigmas = field_outputs[FieldHeadNames.DENSITY]
return rgbs, sigmas
n_rays = ray_bundle.origins.shape[0]
with torch.no_grad():
ray_samples, packed_info = self.sampler(
ray_bundle=ray_bundle,
near_plane=self.config.near_plane,
)
frustum_starts = ray_samples.frustums.starts
frustum_ends = ray_samples.frustums.ends
frustum_dirs = ray_samples.frustums.directions
frustum_positions = ray_samples.frustums.get_positions()
# compat the samples thru volumetric rendering
with torch.no_grad():
densities = self.field.density_fn(frustum_positions)
(
compact_packed_info,
compact_frustum_starts,
compact_frustum_ends,
compact_frustum_positions,
compact_frustum_dirs,
) = nerfacc.volumetric_rendering_steps(
packed_info,
densities,
frustum_starts,
frustum_ends,
frustum_positions,
frustum_dirs,
)
# network
compact_query_results = query_fn(compact_frustum_positions, compact_frustum_dirs)
compact_rgbs, compact_densities = compact_query_results[0], compact_query_results[1]
# accumulation
compact_weights, compact_ray_indices = nerfacc.volumetric_rendering_weights(
compact_packed_info,
compact_densities,
compact_frustum_starts,
compact_frustum_ends,
)
accumulated_color = nerfacc.volumetric_rendering_accumulate(
compact_weights, compact_ray_indices, compact_rgbs, n_rays
)
accumulated_weight = nerfacc.volumetric_rendering_accumulate(compact_weights, compact_ray_indices, None, n_rays)
accumulated_depth = nerfacc.volumetric_rendering_accumulate(
compact_weights,
compact_ray_indices,
(compact_frustum_starts + compact_frustum_ends) / 2.0,
n_rays,
)
if self.config.randomize_background:
# TODO(matt) Investigate randomizing per pixel not per batch.
render_bkgd = torch.rand(3, device=accumulated_color.device)
else:
render_bkgd = torch.ones(3, device=self.device)
accumulated_color = accumulated_color + render_bkgd * (1.0 - accumulated_weight)
alive_ray_mask = accumulated_weight.squeeze(-1) > 0
outputs = {
"rgb": accumulated_color,
"accumulation": accumulated_weight,
"depth": accumulated_depth,
"alive_ray_mask": alive_ray_mask, # the rays we kept from sampler
}
return outputs
def get_loss_dict(self, outputs, batch, metrics_dict, loss_coefficients):
image = batch["image"].to(self.device)
mask = outputs["alive_ray_mask"]
rgb_loss = self.rgb_loss(image[mask], outputs["rgb"][mask])
loss_dict = {"rgb_loss": rgb_loss}
return loss_dict
def get_image_metrics_and_images(
self, outputs: Dict[str, torch.Tensor], batch: Dict[str, torch.Tensor]
) -> Tuple[Dict[str, float], Dict[str, torch.Tensor]]:
image = batch["image"].to(self.device)
rgb = outputs["rgb"]
acc = visualization.apply_colormap(outputs["accumulation"])
depth = visualization.apply_depth_colormap(
outputs["depth"],
accumulation=outputs["accumulation"],
)
alive_ray_mask = visualization.apply_colormap(outputs["alive_ray_mask"])
combined_rgb = torch.cat([image, rgb], dim=1)
combined_acc = torch.cat([acc], dim=1)
combined_depth = torch.cat([depth], dim=1)
combined_alive_ray_mask = torch.cat([alive_ray_mask], dim=1)
# Switch images from [H, W, C] to [1, C, H, W] for metrics computations
image = torch.moveaxis(image, -1, 0)[None, ...]
rgb = torch.moveaxis(rgb, -1, 0)[None, ...]
psnr = self.psnr(image, rgb)
ssim = self.ssim(image, rgb)
lpips = self.lpips(image, rgb)
# all of these metrics will be logged as scalars
metrics_dict = {"psnr": float(psnr.item()), "ssim": float(ssim), "lpips": float(lpips)} # type: ignore
# TODO(ethan): return an image dictionary
images_dict = {
"img": combined_rgb,
"accumulation": combined_acc,
"depth": combined_depth,
"alive_ray_mask": combined_alive_ray_mask,
}
return metrics_dict, images_dict