-
Notifications
You must be signed in to change notification settings - Fork 0
/
visualization.py
122 lines (106 loc) · 5.52 KB
/
visualization.py
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
"""Utility functions for plots and animations."""
import numpy as np
import sys
import matplotlib.pyplot as plt
import matplotlib.animation as animation
from data. data import RawData
import torch
def init_animation(ax, data: RawData, circle: dict={}, number: dict={}) -> dict:
"""Init entities needed in animation and return as a dict.
- "title": Title of the figure
- {ped_id}: The entities corresponding to pedestrian {ped_id}
"""
actors = {}
for ped_id in range(data.num_pedestrians):
actors[ped_id] = {
"circle": plt.Circle((0, 0), **circle, visible=False),
"number": ax.text(0, 0, str(ped_id), **number, size="xx-small", visible=False, verticalalignment="center", horizontalalignment="center", color=(0, 0, 0, 1)),
"legend": ax.text(0.1, 0.9 - 0.08 * ped_id, '', transform=ax.transAxes, visible=False),
"route": (plt.plot([], [], ls='-', marker='.', color=(.5, .5, .5, .1), visible=False))[0],
}
ax.add_patch(actors[ped_id]["circle"])
actors["title"] = plt.title('')
if(data.obstacles.numel()):
plt.plot(data.obstacles[:, 0], data.obstacles[:, 1], "-k")
return actors
def update_animation(frame_num: int, data: RawData, actors: dict, show_speed=False,
color=None) -> list:
frame = data.get_frame(frame_num)
actors_list = []
for ped in range(frame["num_pedestrians"]):
if(frame["mask_p"][ped] == 0):
actors[ped]["circle"].set_visible(False)
actors[ped]["number"].set_visible(False)
actors[ped]["route"].set_visible(False)
if(show_speed):
actors[ped]["legend"].set_visible(False)
continue
speed = np.linalg.norm(frame["velocity"][ped, :])
acc = np.linalg.norm(frame["acceleration"][ped, :])
radius = 0.38 / 2
if(color):
color_ = color(frame)
else:
color_ = (0, 1.34 / (1.34 + speed), speed / (1.34 + speed), 0.4)
actors[ped]["number"].set(position=frame["position"][ped, :], visible=True)
actors[ped]["circle"].set(center=frame["position"][ped, :], radius=radius, color=color_, visible=True)
route = torch.cat((frame['position'][(ped,), :], frame['destinations'][frame['destination_flag'][ped]:, ped, :]), dim=0)
actors[ped]["route"].set(data=(route[:, 0], route[:, 1]), visible=True)
if(show_speed):
actors[ped]["legend"].set(text=f'$v_{{{ped}}} = {speed:.2f}m/s, a_{{{ped}}} = {acc:.2f}m/s^2$', visible=True)
actors_list.append(actors[ped]["circle"])
actors_list.append(actors[ped]["number"])
actors_list.append(actors[ped]["route"])
if(show_speed):
actors_list.append(actors[ped]["legend"])
if("source" in data.meta_data and data.meta_data["source"] == "GC dataset"):
begin_frame = data.meta_data["begin_frame"]
interpolation = data.meta_data["interpolation"]
title_text = f'[GC Dataset]: Frame {int(frame_num//interpolation) + begin_frame} / {frame_num*data.meta_data["time_unit"]:.2f}s'
elif("source" in data.meta_data and data.meta_data["source"] == "basic unit"):
title_text = f'[Basic Unit {data.meta_data["scene"]}]: Frame {frame_num} / {frame_num*data.meta_data["time_unit"]:.2f}s'
else:
title_text = f'Frame {frame_num} / {frame_num*data.meta_data["time_unit"]:.2f}s'
actors["title"].set(text=title_text)
actors_list.append(actors["title"])
return actors_list
def state_animation(ax, data:RawData, *, movie_file=None, writer=None, show_speed=False):
"""Generate animation for {data}."""
if(movie_file): print(f"Saving animation to '{movie_file}'...")
actors = init_animation(ax, data)
def update(i):
progress = round(i / data.num_steps * 100)
print("\r", end="")
print("Animation progress: {}%: ".format(progress), end="")
sys.stdout.flush()
return update_animation(i, data, actors, show_speed)
ani = animation.FuncAnimation(
ax.get_figure(), update,
frames=data.num_steps,
interval=data.meta_data["time_unit"] * 1000.0, blit=True)
if movie_file:
ani.save(movie_file, writer=writer, dpi=200)
return ani
def state_animation_compare(ax, data1:RawData, data2:RawData, *, movie_file=None, writer=None, show_speed=False):
"""Generate animation to compare {data1} and {data2}.
- data1: Data to compare, draw in colorful disks.
- data2: Data as base, draw in black and white circle.
Note: data1 and data2 should have same time unit.
"""
if(movie_file): print(f"Saving compare animation to '{movie_file}'...")
actors1 = init_animation(ax, data1, circle={"zorder":9}, number={"zorder":10})
actors2 = init_animation(ax, data2, circle={"zorder":7}, number={"zorder":8, "alpha":0.2})
def update(i):
progress = round(i / data2.num_steps * 100)
print("\r", end="")
print("Animation progress: {}%: ".format(progress), end="")
sys.stdout.flush()
return update_animation(i, data1, actors1, show_speed) \
+ update_animation(i, data2, actors2, show_speed, color=lambda x: (0.2, 0.2, 0.2, 0.2))
ani = animation.FuncAnimation(
ax.get_figure(), update,
frames=data2.num_steps,
interval=data2.meta_data["time_unit"] * 1000.0, blit=True)
if movie_file:
ani.save(movie_file, writer=writer, dpi=200)
return ani