jgrewe/fixtracks
1
0
Fork 0
Code Issues 4 Pull Requests Actions Packages Projects Releases Wiki Activity

working on merging

Browse Source
This commit is contained in:
Jan Grewe 2024-12-30 13:46:16 +01:00
parent b24c4435e7
commit 39dadac57e
4 changed files with 258 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
README.md
View File

@ -1,3 +1,20 @@
# fitracks
UI tool to merge and fix tracking results from our 2-camera behavior setup
UI tool to merge and fix tracking results from our 2-camera behavior setup
## TODO
* add status bar that shows the processes that are running in the background
* icons
* merging of the detections
* track merging
* leftcut: 1778, rightcut: 400
* how to merge the detections:
1. go through the frames.
2. for each frame:
1. get the detections for that frame
2. if they are outside the overlap range, accept them into the destination data frame
* if on the right video, increment the x-positions
* if they are on the left video, accept as is
3. if they are within, they need to be merged:

81
fixtracks/detectionmerge.py
View File

@ -1,10 +1,10 @@
import logging
from PyQt6.QtWidgets import QWidget, QGridLayout, QVBoxLayout, QHBoxLayout, QLabel, QPushButton, QComboBox, QSizePolicy, QSpinBox, QGraphicsView, QGraphicsScene, QGraphicsLineItem, QSpacerItem
from PyQt6.QtCore import QThreadPool, Qt, pyqtSignal
from PyQt6.QtWidgets import QWidget, QGridLayout, QVBoxLayout, QHBoxLayout, QLabel, QPushButton, QComboBox, QSizePolicy, QSpinBox, QGraphicsView, QGraphicsScene, QGraphicsLineItem, QSpacerItem, QProgressDialog, QProgressBar
from PyQt6.QtCore import QThreadPool, Qt, pyqtSignal, pyqtSlot
from PyQt6.QtGui import QImage, QPixmap, QColor, QPen
from fixtracks.util import ImageReader, DataFrameReader
from fixtracks.util import ImageReader, DataFrameReader, Merger
class VideoPreview(QWidget):
@ -59,6 +59,8 @@ class MergeDetections(QWidget):
self.left_dataframereader = None
self._left_data = None
self._right_data = None
self._progressDialog = None
self._merger = None
self.left_datacombo = QComboBox()
self.left_videocombo = QComboBox()
@ -67,6 +69,9 @@ class MergeDetections(QWidget):
self.left_posspinner = QSpinBox()
self.left_posspinner.valueChanged.connect(self.on_leftmergelinemove)
self.left_preview = VideoPreview()
self.left_framespinner = QSpinBox()
self.left_framespinner.setMaximum(10000)
self.left_framespinner.setValue(100)
self.right_datacombo = QComboBox()
self.right_videocombo = QComboBox()
@ -75,6 +80,9 @@ class MergeDetections(QWidget):
self.right_posspinner = QSpinBox()
self.right_posspinner.valueChanged.connect(self.on_rightmergelinemove)
self.right_preview = VideoPreview(left=False)
self.right_framespinner = QSpinBox()
self.right_framespinner.setMaximum(10000)
self.right_framespinner.setValue(100)
self._mergePreviewBtn = QPushButton("Preview")
self._mergePreviewBtn.clicked.connect(self.on_mergePreview)
@ -112,17 +120,21 @@ class MergeDetections(QWidget):
grd.addWidget(QLabel("Data"), 1, 0)
grd.addWidget(QLabel("Video"), 2, 0)
grd.addWidget(QLabel("Mergeline"), 3, 0)
grd.addWidget(QLabel("Seek frame"), 4, 0)
grd.addWidget(self.left_datacombo, 1, 1)
grd.addWidget(self.left_videocombo, 2, 1)
grd.addWidget(self.left_posspinner, 3, 1)
grd.addWidget(self.left_framespinner, 4, 1)
grd.addWidget(QLabel("Right"), 0, 2)
grd.addWidget(QLabel("Data"), 1, 2)
grd.addWidget(QLabel("Video"), 2, 2)
grd.addWidget(QLabel("Mergeline"), 3, 2)
grd.addWidget(QLabel("Seek frame"), 4, 2)
grd.addWidget(self.right_datacombo, 1, 3)
grd.addWidget(self.right_videocombo, 2, 3)
grd.addWidget(self.right_posspinner, 3, 3)
grd.addWidget(self.right_framespinner, 4, 3)
grd.setColumnStretch(0, 0)
grd.setColumnStretch(2, 0)
@ -173,33 +185,39 @@ class MergeDetections(QWidget):
def on_rightvideoSelection(self):
logging.debug("Video selection of the %s side", "right")
self.right_imagereader = ImageReader(self.right_videocombo.currentText(), 100)
self.right_imagereader = ImageReader(self.right_videocombo.currentText(), self.right_framespinner.value())
self.right_imagereader.signals.finished.connect(self.right_imgreaderDone)
self.threadpool.start(self.right_imagereader)
def right_imgreaderDone(self, state):
logging.debug("Right image reader done with state %s", str(state))
frame = self.right_imagereader.frame
img = self._toImage(frame)
self.right_preview.set_image(img)
self.right_posspinner.setMaximum(img.width() - 1)
self.right_posspinner.setValue(0)
self.checkButtons()
if frame is not None:
img = self._toImage(frame)
self.right_preview.set_image(img)
self.right_posspinner.setMaximum(img.width() - 1)
self.right_posspinner.setValue(0)
self.checkButtons()
else:
logging.error("Reading frame failed!")
def on_leftVideoSelection(self):
logging.debug("Video selection of the %s side", "left")
self.left_imagereader = ImageReader(self.left_videocombo.currentText(), 100)
self.left_imagereader = ImageReader(self.left_videocombo.currentText(), self.left_framespinner.value())
self.left_imagereader.signals.finished.connect(self.left_imgreaderDone)
self.threadpool.start(self.left_imagereader)
def left_imgreaderDone(self, state):
logging.debug("Left image reader done with state %s", str(state))
frame = self.left_imagereader.frame
img = self._toImage(frame)
self.left_preview.set_image(img)
self.left_posspinner.setMaximum(img.width() - 1)
self.left_posspinner.setValue(img.width() - 1)
self.checkButtons()
if frame is not None:
img = self._toImage(frame)
self.left_preview.set_image(img)
self.left_posspinner.setMaximum(img.width() - 1)
self.left_posspinner.setValue(img.width() - 1)
self.checkButtons()
else:
logging.error("Reading frame failed!")
def on_leftDataSelection(self):
logging.debug("Data selection of the %s side", "left")
@ -241,8 +259,37 @@ class MergeDetections(QWidget):
def on_merge(self):
logging.debug("detectionmerge: merge pressed")
pass
self.merger = Merger(self._left_data, self._right_data,
self.left_posspinner.value(),
self.right_posspinner.value())
self._progressDialog = QProgressDialog(parent=self)
self._progressDialog.setAutoClose(True)
self._progressDialog.setRange(0, 100)
self._progressDialog.setLabelText("Merging detections:")
self._progressDialog.setCancelButtonText("Cancel")
self._progressDialog.setWindowModality(Qt.WindowModality.WindowModal)
self._progressDialog.canceled.connect(self.on_mergeCancelled)
self._progressDialog.show()
self.merger.signals.progress.connect(self.on_mergeProgress)
self.merger.signals.finished.connect(self.on_mergeDone)
self.threadpool.start(self.merger)
@pyqtSlot()
def on_mergeCancelled(self):
self.merger.stop_request()
@pyqtSlot(float)
def on_mergeProgress(self, value):
print("mergeProgress", value)
if self._progressDialog is not None:
self._progressDialog.setValue(int(value * 100))
@pyqtSlot(bool)
def on_mergeDone(self, state):
logging.debug("Merging stopped with status %s", state)
self._progressDialog.close()
self._progressDialog = None
self.merger = None
def checkButtons(self):
merge_enabled = self._left_data is not None and self._right_data is not None

11
fixtracks/taskwidget.py
View File

@ -16,6 +16,13 @@ class TasksWidget(QWidget):
folderBtn.setMaximumSize(200, 100)
folderBtn.clicked.connect(self._open_folder)
self.convertBtn = QPushButton("Convert tracks")
self.convertBtn.setToolTip("Convert JSON pose files to to Pandas data frames")
self.convertBtn.setEnabled(False)
self.convertBtn.setMaximumSize(200, 100)
self.convertBtn.clicked.connect(self.convert_clicked)
self.convertBtn.setSizePolicy(QSizePolicy.Policy.Expanding, QSizePolicy.Policy.Expanding)
self.mergeBtn = QPushButton("Merge detections")
self.mergeBtn.setEnabled(False)
self.mergeBtn.setMaximumSize(200, 100)
@ -31,6 +38,7 @@ class TasksWidget(QWidget):
l = QVBoxLayout()
l.setAlignment(Qt.AlignmentFlag.AlignVCenter)
l.addWidget(folderBtn)
l.addWidget(self.convertBtn)
l.addWidget(self.mergeBtn)
l.addWidget(self.tracksBtn)
self.setLayout(l)
@ -42,6 +50,9 @@ class TasksWidget(QWidget):
def _tracks_clicked(self):
self.clicked.emit("Tracks")
def _convert_clickes(self)
self.clicked.emit("Convert")
def _open_folder(self):
logging.debug("TasksWidget:select data folder")
folder = QFileDialog.getExistingDirectory()

167
fixtracks/util.py
View File

@ -1,14 +1,18 @@
import logging
import pandas as pd
import cv2 as cv
import time
import numpy as np
from PyQt6.QtCore import QRunnable, pyqtSlot, pyqtSignal, QObject
from IPython import embed
class ProducerSignals(QObject):
finished = pyqtSignal(bool)
# error = pyqtSignal(str)
# start = pyqtSignal(float)
# running = pyqtSignal()
progress = pyqtSignal(float)
class ImageReader(QRunnable):
@ -28,6 +32,9 @@ class ImageReader(QRunnable):
'''
logging.debug("ImageReader: trying to open file %s", self._filename)
cap = cv.VideoCapture(self._filename)
framecount = int(cap.get(cv.CAP_PROP_FRAME_COUNT))
if self._framenumber >= framecount:
logging.warning("ImageReader: desired frame number (%i) exceeds the frame count (%i)! Redefined to frame count." % (self._framenumber, framecount))
if not cap.isOpened():
logging.debug("ImageReader: failed to open file %s", self._filename)
self._signals.finished.emit(False)
@ -39,7 +46,8 @@ class ImageReader(QRunnable):
self._signals.finished.emit(False)
break
fn += 1
self._frame = frame # cv.cvtColor(frame, cv.COLOR_BGR2RGB)
self._frame = frame # cv.cvtColor(frame, cv.COLOR_BGR2RGB)
cap.release()
self._signals.finished.emit(True)
@property
@ -77,3 +85,158 @@ class DataFrameReader(QRunnable):
def dataframe(self):
return self._dataframe
class Merger(QRunnable):
def __init__(self, left_data, right_data, left_cut, right_cut) -> None:
super().__init__()
self._signals = ProducerSignals()
self._left_data = left_data
self._right_data = right_data
self._left_cut = left_cut
self._right_cut = right_cut
self._result = None
self._stopRequest = False
@pyqtSlot()
def stop_request(self):
self._stopRequest = True
@pyqtSlot()
def run(self):
logging.debug("Merger: running merge for %i frames", max_frames)
self._stopRequest = False
max_frames = max(self._left_data.frame.max(), self._right_data.frame.max())
step = max_frames // 100
self._result = pd.DataFrame(columns=self._left_data.columns)
for frame in range(max_frames):
if self._stopRequest:
break
lf = self._left_data[self._left_data.frame == frame]
rf = self._right_data[self._right_data.frame == frame]
merge_frame(lf, rf, self._left_cut, self._right_cut, self._result)
if frame % step == 0:
self.signals.progress.emit(frame/max_frames)
time.sleep(0.01)
self._signals.finished.emit(True and (not self._stopRequest))
@property
def signals(self):
return self._signals
@property
def result(self):
return self._result
def merge_frames(left, right, leftcut, rightcut, destination):
# for
pass
def check_frame(frame, cut, left=True):
"""checks whether the detected object is (partially) in the overlapping zone of the two cameras.
A frame is 'ok' if the box is not in the danger zone.
Parameters
----------
frame : pd.Series,
a row of the DataFrame
cut : int
The cut x-position
left : bool, optional
whether we are looking at the right or left frame, by default True
Returns
-------
bool
whether or not the
"""
if left:
return not (frame.box_x1 > cut or frame.box_x2 > cut) # any of the box coordinates is beyond the cut
else:
return not (frame.box_x1 < cut or frame.box_x2 < cut)
def merge_detections(left_data:pd.DataFrame, right_data: pd.DataFrame, left_cut: int, right_cut: int) ->pd.DataFrame:
"""Merge the key-point detections based on the left and right video. Key-points with x-coordinates beyond each of the limits (left and right cut) are discarded.
How to merge the detections WITHIN one of the fish detections???
Parameters
----------
left_data : pd.DataFrame
Detections based on the left video
right_data : pd.DataFrame
Detections based on the right video.
left_cut : int
Where to cut off and discard the detected key points.
right_cut : int
Where to cut off and discard the detected key points.
Returns
-------
pd.DataFrame
merged detections of left and right data
"""
def check_dataframe(df):
return True
def to_numpy(df):
key_columns = [c for c in df.columns if "key_" in c]
box_columns = [c for c in df.columns if "box_" in c]
num_frames = len(df)
num_keypoints = len(key_columns)
dimensions = 2
keypoints = np.empty((num_frames, num_keypoints, dimensions))
visibility = np.empty((num_frames, num_keypoints))
boxcoordinates = np.empty((num_frames, 4))
for i, row in df.iterrows():
for j, k in enumerate(key_columns):
key_data = row[k]
l = list(map(float, list(key_data[1:-1].split(","))))
keypoints[i, j, :] = l
for j, b in enumerate(box_columns):
boxcoordinates[i, j] = row[b]
if isinstance(row["visible"], str):
vis = list(map(float, row["visible"][1:-1].split()))
visibility[i, :] = vis
else:
visibility[i, :] = row["visible"]
return keypoints, visibility, boxcoordinates
logging.debug("Cutting left detections to limit %i", left_cut)
if not check_dataframe(left_data) or not check_dataframe(right_data):
logging.error("Left or right dataframe structure does not match my expectations")
return None
df = pd.DataFrame(columns=left_data.columns)
dangerzone = {"left": [], "right":[]}
lkeypoints, lquality, lbox = to_numpy(left_data)
embed()
exit()
for i, row in left_data.iterrows():
if check_frame(row, left_cut):
df = pd.concat([df, row.to_frame().T], ignore_index=True)
else:
dangerzone["left"].append(row)
if i > 10000:
break
print("Left done")
for i, row in right_data.iterrows():
if check_frame(row[1], right_cut, left=False):
# convert_right(row[1]) need to convert the frame coordinates!
df = pd.concat([df, row[1].to_frame().T], ignore_index=True)
else:
dangerzone["right"].append(row[1])
if i > 10000:
break
def main():
left = pd.read_csv("../data/left_tracks.csv", sep=";", index_col=0)
# right = pd.read_csv("../data/right_tracks.csv", sep=";", index_col=0)
merge_detections(left, None, 2000, 300)
if __name__ == "__main__":
main()