aboutsummaryrefslogtreecommitdiff
path: root/src/argaze/GazeAnalysis/VelocityThresholdIdentification.py
blob: 7d6c7b2f2968772c8399e90002637573e95225b8 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
#!/usr/bin/env python

"""Implementation of the I-VT algorithm as described in:
        
    **Dario D. Salvucci and Joseph H. Goldberg (2000).**  
    *Identifying fixations and saccades in eye-tracking protocols.*  
    In Proceedings of the 2000 symposium on Eye tracking research & applications (ETRA'00, 71-78).  
    [https://doi.org/10.1145/355017.355028](https://doi.org/10.1145/355017.355028)
"""

__author__ = "Théo de la Hogue"
__credits__ = []
__copyright__ = "Copyright 2023, Ecole Nationale de l'Aviation Civile (ENAC)"
__license__ = "BSD"

from typing import TypeVar, Tuple
from dataclasses import dataclass, field
import math

from argaze import GazeFeatures

import numpy
import cv2

GazeMovementType = TypeVar('GazeMovement', bound="GazeMovement")
# Type definition for type annotation convenience

FixationType = TypeVar('Fixation', bound="Fixation")
# Type definition for type annotation convenience

SaccadeType = TypeVar('Saccade', bound="Saccade")
# Type definition for type annotation convenience

@dataclass(frozen=True)
class Fixation(GazeFeatures.Fixation):
    """Define dispersion based fixation."""

    deviation_max: float = field(init=False)
    """Maximal gaze position distance to the centroïd."""

    def __post_init__(self):

        super().__post_init__()

        points = self.positions.values()
        points_x, points_y = [p[0] for p in points], [p[1] for p in points]
        points_array = numpy.column_stack([points_x, points_y])
        centroid_array = numpy.array([numpy.mean(points_x), numpy.mean(points_y)])
        deviations_array = numpy.sqrt(numpy.sum((points_array - centroid_array)**2, axis=1))

        # Update frozen focus attribute using centroid
        object.__setattr__(self, 'focus', (centroid_array[0], centroid_array[1]))

        # Update frozen deviation_max attribute
        object.__setattr__(self, 'deviation_max', max(deviations_array))

    def point_deviation(self, gaze_position) -> float:
        """Get distance of a point from the fixation's centroïd."""

        return numpy.sqrt((self.centroid[0] - gaze_position.value[0])**2 + (self.centroid[1] - gaze_position.value[1])**2)

    def overlap(self, fixation) -> bool:
        """Does a gaze position from another fixation having a deviation to this fixation centroïd smaller than maximal deviation?"""

        points = fixation.positions.values()
        points_x, points_y = [p[0] for p in points], [p[1] for p in points]
        points_array = numpy.column_stack([points_x, points_y])
        centroid_array = numpy.array([self.centroid[0], self.centroid[1]])
        deviations_array = numpy.sqrt(numpy.sum((points_array - centroid_array)**2, axis=1))

        return min(deviations_array) <= self.deviation_max

    def merge(self, fixation) -> FixationType:
        """Merge another fixation into this fixation."""

        self.positions.append(fixation.positions)
        self.__post_init__()

        return self

    def draw(self, image: numpy.array, color=(127, 127, 127), border_color=(255, 255, 255)):
        """Draw fixation into image."""

        cv2.circle(image, (int(self.focus[0]), int(self.focus[1])), int(self.deviation_max), color, -1)
        cv2.circle(image, (int(self.focus[0]), int(self.focus[1])), int(self.deviation_max), border_color, len(self.positions))

@dataclass(frozen=True)
class Saccade(GazeFeatures.Saccade):
    """Define dispersion based saccade."""

    def __post_init__(self):
        super().__post_init__()

    def draw(self, image: numpy.array, color=(255, 255, 255)):
        """Draw saccade into image."""

        _, start_position = self.positions.first
        _, last_position = self.positions.last

        cv2.line(image, (int(start_position[0]), int(start_position[1])), (int(last_position[0]), int(last_position[1])), color, 2)

@dataclass
class GazeMovementIdentifier(GazeFeatures.GazeMovementIdentifier):
    """Implementation of the I-VT algorithm as described in:
        
            Dario D. Salvucci and Joseph H. Goldberg. 2000. Identifying fixations and
            saccades in eye-tracking protocols. In Proceedings of the 2000 symposium
            on Eye tracking research & applications (ETRA '00). ACM, New York, NY, USA,
            71-78. [http://dx.doi.org/10.1145/355017.355028](http://dx.doi.org/10.1145/355017.355028)
    """

    velocity_max_threshold: int|float
    """Maximal velocity allowed to consider a gaze movement as a fixation."""

    duration_min_threshold: int|float
    """Minimal duration allowed to consider a gaze movement as a fixation.
    It is also used as maximal duration allowed to consider a gaze movement as a saccade."""

    def __post_init__(self):

        self.__last_ts = -1
        self.__last_position = None

        self.__fixation_positions = GazeFeatures.TimeStampedGazePositions()
        self.__saccade_positions = GazeFeatures.TimeStampedGazePositions()

    def identify(self, ts, gaze_position, terminate=False) -> GazeMovementType:
        """Identify gaze movement from successive timestamped gaze positions.

            The optional *terminate* argument allows to notify identification algorithm that given gaze position will be the last one.
        """

        # Ignore non valid gaze position
        if not gaze_position.valid:

            return None if not terminate else self.current_fixation

        # Store first valid position
        if self.__last_ts < 0:

            self.__last_ts = ts
            self.__last_position = gaze_position

            return

        # Check if too much time elapsed since last gaze position
        if (ts - self.__last_ts) > self.duration_min_threshold:

            # Remember last position
            self.__last_ts = ts
            self.__last_position = gaze_position

            # Get last movement
            last_movement = self.current_saccade if len(self.__fixation_positions) == 0 else self.current_fixation

            # Clear all former gaze positions
            self.__fixation_positions = GazeFeatures.TimeStampedGazePositions()
            self.__saccade_positions = GazeFeatures.TimeStampedGazePositions()

            # Return last valid movement if exist
            return last_movement
        
        # Velocity
        velocity = abs(gaze_position.distance(self.__last_position) / (ts - self.__last_ts))

        # Remember last position
        self.__last_ts = ts
        self.__last_position = gaze_position

        # Velocity is greater than threshold
        if velocity > self.velocity_max_threshold:

            # Append to saccade positions
            self.__saccade_positions[ts] = gaze_position

            # Does last fixation exist?
            if len(self.__fixation_positions) > 0:

                last_fixation = Fixation(self.__fixation_positions)

                # Clear fixation positions
                self.__fixation_positions = GazeFeatures.TimeStampedGazePositions()

                # Output last fixation
                return last_fixation

            # Identification must stop: ends with current saccade
            if terminate:

                return self.current_saccade

        # Velocity is less or equals to threshold
        else:

            # Append to fixation positions
            self.__fixation_positions[ts] = gaze_position

            # Does last saccade exist?
            if len(self.__saccade_positions) > 0:

                last_saccade = Saccade(self.__saccade_positions)

                # Clear fixation positions
                self.__saccade_positions = GazeFeatures.TimeStampedGazePositions()

                # Output last saccade
                return last_saccade

            # Identification must stop: ends with current fixation
            if terminate:

                return self.current_fixation

    @property
    def current_fixation(self) -> FixationType:

        if len(self.__fixation_positions) > 0:

            return Fixation(self.__fixation_positions)

    @property
    def current_saccade(self) -> SaccadeType:

        if len(self.__saccade_positions) > 0:

            return Saccade(self.__saccade_positions)