Filtering outlier input cameras in trajectory estimation

Summary: Useful for visualising colmap output where some frames are not correctly registered.

Reviewed By: bottler

Differential Revision: D38743191

fbshipit-source-id: e823df2997870dc41d76784e112d4349f904d311

pytorch3d/implicitron/tools/eval_video_trajectory.py

@@ -4,16 +4,21 @@
 # This source code is licensed under the BSD-style license found in the
 # LICENSE file in the root directory of this source tree.
 
+import logging
 import math
 from typing import Optional, Tuple
 
 import torch
 from pytorch3d.common.compat import eigh
+from pytorch3d.implicitron.tools import utils
 from pytorch3d.implicitron.tools.circle_fitting import fit_circle_in_3d
 from pytorch3d.renderer import look_at_view_transform, PerspectiveCameras
 from pytorch3d.transforms import Scale
 
 
+logger = logging.getLogger(__name__)
+
+
 def generate_eval_video_cameras(
     train_cameras,
     n_eval_cams: int = 100,
@@ -27,6 +32,7 @@ def generate_eval_video_cameras(
     infer_up_as_plane_normal: bool = True,
     traj_offset: Optional[Tuple[float, float, float]] = None,
     traj_offset_canonical: Optional[Tuple[float, float, float]] = None,
+    remove_outliers_rate: float = 0.0,
 ) -> PerspectiveCameras:
     """
     Generate a camera trajectory rendering a scene from multiple viewpoints.
@@ -50,9 +56,16 @@ def generate_eval_video_cameras(
             Active for the `trajectory_type="circular"`.
         scene_center: The center of the scene in world coordinates which all
             the cameras from the generated trajectory look at.
+        remove_outliers_rate: the number between 0 and 1; if > 0,
+            some outlier train_cameras will be removed from trajectory estimation;
+            the filtering is based on camera center coordinates; top and
+            bottom `remove_outliers_rate` cameras on each dimension are removed.
     Returns:
         Dictionary of camera instances which can be used as the test dataset
     """
+    if remove_outliers_rate > 0.0:
+        train_cameras = _remove_outlier_cameras(train_cameras, remove_outliers_rate)
+
     if trajectory_type in ("figure_eight", "trefoil_knot", "figure_eight_knot"):
         cam_centers = train_cameras.get_camera_center()
         # get the nearest camera center to the mean of centers
@@ -167,6 +180,20 @@ def generate_eval_video_cameras(
     return test_cameras
 
 
+def _remove_outlier_cameras(
+    cameras: PerspectiveCameras, outlier_rate: float
+) -> PerspectiveCameras:
+    keep_indices = utils.get_inlier_indicators(
+        cameras.get_camera_center(), dim=0, outlier_rate=outlier_rate
+    )
+    clean_cameras = cameras[keep_indices]
+    logger.info(
+        "Filtered outlier cameras when estimating the trajectory: "
+        f"{len(cameras)} → {len(clean_cameras)}"
+    )
+    return clean_cameras
+
+
 def _disambiguate_normal(normal, up):
     up_t = torch.tensor(up).to(normal)
     flip = (up_t * normal).sum().sign()

pytorch3d/implicitron/tools/utils.py

@@ -9,7 +9,7 @@
 import dataclasses
 import time
 from contextlib import contextmanager
-from typing import Any, Callable, Dict
+from typing import Any, Callable, Dict, Iterable, Iterator
 
 import torch
 
@@ -157,6 +157,26 @@ def cat_dataclass(batch, tensor_collator: Callable):
     return type(elem)(**collated)
 
 
+def recursive_visitor(it: Iterable[Any]) -> Iterator[Any]:
+    for x in it:
+        if isinstance(x, Iterable) and not isinstance(x, (str, bytes)):
+            yield from recursive_visitor(x)
+        else:
+            yield x
+
+
+def get_inlier_indicators(
+    tensor: torch.Tensor, dim: int, outlier_rate: float
+) -> torch.Tensor:
+    remove_elements = int(min(outlier_rate, 1.0) * tensor.shape[dim] / 2)
+    hi = torch.topk(tensor, remove_elements, dim=dim).indices.tolist()
+    lo = torch.topk(-tensor, remove_elements, dim=dim).indices.tolist()
+    remove_indices = set(recursive_visitor([hi, lo]))
+    keep_indices = tensor.new_ones(tensor.shape[dim : dim + 1], dtype=torch.bool)
+    keep_indices[list(remove_indices)] = False
+    return keep_indices
+
+
 class Timer:
     """
     A simple class for timing execution.