fix #2318: add train/test/eval split for instant ngp dataparser (#2335)

* fix #2318: add train/test/eval split for instant ngp dataparser

* fix: instant_ngp_dataparser.py format
This commit is contained in:
elenacliu authored and GitHub committed 2023-08-14 04:24:34 +00:00
1 parent 3455fb749f
commit bc8341378a
1 file changed
+48 -2
@@ -18,7 +18,7 @@ from __future__ import annotations
from dataclasses import dataclass, field
from pathlib import Path
from typing import Dict, Tuple, Type
from typing import Dict, Tuple, Type, Literal
import imageio
import numpy as np
@@ -32,6 +32,12 @@ from nerfstudio.data.dataparsers.base_dataparser import (
DataparserOutputs,
)
from nerfstudio.data.scene_box import SceneBox
from nerfstudio.data.utils.dataparsers_utils import (
get_train_eval_split_filename,
get_train_eval_split_fraction,
get_train_eval_split_interval,
get_train_eval_split_all,
)
from nerfstudio.utils.io import load_from_json
from nerfstudio.utils.rich_utils import CONSOLE
@@ -46,6 +52,18 @@ class InstantNGPDataParserConfig(DataParserConfig):
"""Directory or explicit json file path specifying location of data."""
scene_scale: float = 0.3333
"""How much to scale the scene."""
eval_mode: Literal["fraction", "filename", "interval", "all"] = "fraction"
"""
The method to use for splitting the dataset into train and eval.
Fraction splits based on a percentage for train and the remaining for eval.
Filename splits based on filenames containing train/eval.
Interval uses every nth frame for eval.
All uses all the images for any split.
"""
train_split_fraction: float = 0.9
"""The percentage of the dataset to use for training. Only used when eval_mode is train-split-fraction."""
eval_interval: int = 8
"""The interval between frames to use for eval. Only used when eval_mode is eval-interval."""
@dataclass
@@ -94,12 +112,40 @@ class InstantNGP(DataParser):
assert (
len(image_filenames) != 0
), """
No image files found.
No image files found.
You should check the file_paths in the transforms.json file to make sure they are correct.
"""
poses = np.array(poses).astype(np.float32)
poses[:, :3, 3] *= self.config.scene_scale
# find train and eval indices based on the eval_mode specified
if self.config.eval_mode == "fraction":
i_train, i_eval = get_train_eval_split_fraction(image_filenames, self.config.train_split_fraction)
elif self.config.eval_mode == "filename":
i_train, i_eval = get_train_eval_split_filename(image_filenames)
elif self.config.eval_mode == "interval":
i_train, i_eval = get_train_eval_split_interval(image_filenames, self.config.eval_interval)
elif self.config.eval_mode == "all":
CONSOLE.log(
"[yellow] Be careful with '--eval-mode=all'. If using camera optimization, the cameras may diverge in the current implementation, giving unpredictable results."
)
i_train, i_eval = get_train_eval_split_all(image_filenames)
else:
raise ValueError(f"Unknown eval mode {self.config.eval_mode}")
if split == "train":
indices = i_train
elif split in ["val", "test"]:
indices = i_eval
else:
raise ValueError(f"Unknown dataparser split {split}")
# Choose image_filenames and poses based on split, but after auto orient and scaling the poses.
image_filenames = [image_filenames[i] for i in indices]
mask_filenames = [mask_filenames[i] for i in indices] if len(mask_filenames) > 0 else []
idx_tensor = torch.tensor(indices, dtype=torch.long)
poses = poses[idx_tensor]
camera_to_world = torch.from_numpy(poses[:, :3]) # camera to world transform
distortion_params = camera_utils.get_distortion_params(