Visitar URL original
Make 3D Triangulate possible with pytorch_dlc branch and MA project by Zelin2001 · Pull Request #2729 · DeepLabCut/DeepLabCut · GitHub
Skip to content

Make 3D Triangulate possible with pytorch_dlc branch and MA project #2729

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
MMathisLab merged 3 commits into from
Sep 9, 2024
Merged

Make 3D Triangulate possible with pytorch_dlc branch and MA project #2729

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3e75f90
Bugfix: triangulation
Zelin2001 Sep 2, 2024
dbacba3
use analyze_videos from the compact module
Zelin2001 Sep 7, 2024
37f8477
remove logging lines
Zelin2001 Sep 7, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
62 changes: 36 additions & 26 deletions deeplabcut/pose_estimation_3d/triangulation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,25 +4,21 @@
# https://github.com/DeepLabCut/DeepLabCut
#
# Please see AUTHORS for contributors.
# https://github.com/DeepLabCut/DeepLabCut/blob/master/AUTHORS
# https://github.com/DeepLabCut/DeepLabCut/blob/main/AUTHORS
#
# Licensed under GNU Lesser General Public License v3.0
#

import os
from pathlib import Path

import cv2
import numpy as np
import pandas as pd
from matplotlib.axes._axes import _log as matplotlib_axes_logger

from deeplabcut.utils import auxfun_multianimal, auxiliaryfunctions
from deeplabcut.utils import auxiliaryfunctions_3d
from deeplabcut.core.trackingutils import TRACK_METHODS

matplotlib_axes_logger.setLevel("ERROR")


def triangulate(
config,
Expand Down Expand Up @@ -88,7 +84,7 @@ def triangulate(
To analyze only a few pair of videos:
>>> deeplabcut.triangulate(config,[['C:\\yourusername\\rig-95\\Videos\\video1-camera-1.avi','C:\\yourusername\\rig-95\\Videos\\video1-camera-2.avi'],['C:\\yourusername\\rig-95\\Videos\\video2-camera-1.avi','C:\\yourusername\\rig-95\\Videos\\video2-camera-2.avi']])
"""
from deeplabcut.pose_estimation_tensorflow import predict_videos
from deeplabcut.compat import analyze_videos
from deeplabcut.post_processing import filtering

cfg_3d = auxiliaryfunctions.read_config(config)
Expand Down Expand Up @@ -263,7 +259,7 @@ def triangulate(
scorer_name[cam_names[j]] = DLCscorer
else:
# Analyze video if score name is different
DLCscorer = predict_videos.analyze_videos(
DLCscorer = analyze_videos(
config_2d,
[video],
videotype=videotype,
Expand Down Expand Up @@ -293,7 +289,7 @@ def triangulate(
)

else: # need to do the whole jam.
DLCscorer = predict_videos.analyze_videos(
DLCscorer = analyze_videos(
config_2d,
[video],
videotype=videotype,
Expand Down Expand Up @@ -446,18 +442,32 @@ def triangulate(
}

# Create 3D DataFrame column and row indices
axis_labels = ("x", "y", "z")
cols = [
[scorer_3d],
list(auxiliaryfunctions.get_bodyparts(cfg)),
["x", "y", "z"],
]
cols_names = ["scorer", "bodyparts", "coords"]
flag_indiv_single = False
if cfg.get("multianimalproject"):
columns = pd.MultiIndex.from_product(
[[scorer_3d], individuals, bodyparts, axis_labels],
names=["scorer", "individuals", "bodyparts", "coords"],
)

else:
columns = pd.MultiIndex.from_product(
[[scorer_3d], bodyparts, axis_labels],
names=["scorer", "bodyparts", "coords"],
)
cols_names.insert(1, "individuals")
if "single" == individuals[-1]:
individuals = individuals[:-1]
columns_unique = pd.MultiIndex.from_product(
[
[scorer_3d],
["single"],
auxiliaryfunctions.get_unique_bodyparts(cfg),
["x", "y", "z"],
],
names=cols_names,
)
flag_indiv_single = True
cols.insert(1, individuals)
columns = pd.MultiIndex.from_product(cols, names=cols_names)
if flag_indiv_single:
columns = columns.append(columns_unique)
individuals.append("single")

inds = range(num_frames)

Expand All @@ -468,10 +478,10 @@ def triangulate(
df_3d = pd.DataFrame(triangulate, columns=columns, index=inds)

df_3d.to_hdf(
str(output_filename + ".h5"),
"df_with_missing",
format="table",
str(output_filename) + ".h5",
key="df_with_missing",
mode="w",
format="table",
)

# Reorder 2D dataframe in view 2 to match order of view 1
Expand All @@ -483,19 +493,19 @@ def triangulate(
)
df_2d_view2.to_hdf(
dataname[1],
"tracks",
key="tracks",
format="table",
mode="w",
)

auxiliaryfunctions_3d.SaveMetadata3d(
str(output_filename + "_meta.pickle"), metadata
str(output_filename) + "_meta.pickle", metadata
)

if save_as_csv:
df_3d.to_csv(str(output_filename + ".csv"))
df_3d.to_csv(str(output_filename) + ".csv")

print("Triangulated data for video", video_list[i])
print("Triangulated data for video", video)
print("Results are saved under: ", destfolder)
# have to make the dest folder none so that it can be updated for a new pair of videos
if destfolder == str(Path(video).parents[0]):
Expand Down
15 changes: 10 additions & 5 deletions deeplabcut/utils/auxiliaryfunctions_3d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -322,12 +322,17 @@ def _associate_paired_view_tracks(tracklets1, tracklets2, F):
_t1 = np.c_[_t1, np.ones((*_t1.shape[:2], 1))]
_t2 = np.c_[_t2, np.ones((*_t2.shape[:2], 1))]

# cost for any point in time of t1 being the same
# any point in time of t2
cost = np.abs(np.nansum(np.matmul(_t1, F) * _t2, axis=2))
try:
# cost for any point in time of t1 being the same
# any point in time of t2
cost = np.abs(np.nansum(np.matmul(_t1, F) * _t2, axis=2))

# Get average cost of the entire track
cost = cost.mean()
except:
# typically when dim 2 differs, with uniquebodyparts
cost = 100000.0

# Get average cost of the entire track
cost = cost.mean()
costs[i, j] = cost

match_inds = linear_sum_assignment(np.abs(costs))
Expand Down