-
Notifications
You must be signed in to change notification settings - Fork 0
/
config.yaml
133 lines (130 loc) · 3.93 KB
/
config.yaml
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
#framerate
framerate: 30
#depth maximum value, 0 is the minimum
max_depth: 12000
#depth intensity, above which everything is considered noise
noise_thres: 0.35
#read=d/f/k; read from data/file/kinect
read: k
#if read from kinect, live or recorded
stream: live
#if recorded stream, the .bag path
bag_path: /media/vassilis/Thesis/Datasets/PersonalFarm/rosbag_files/data.bag
#save=y/n
save: n
#directories to load images/test data
path_depth: /media/vassilis/Thesis/Datasets/PersonalFarm/MovingObject/Depth
path_color: /media/vassilis/Thesis/Datasets/PersonalFarm/MovingObject/Color
#directories to load train data
actions_path: /media/vassilis/Thesis/Datasets/PersonalFarm/actions/sets
mv_obj_fold_name : mv_obj
hnd_mk_fold_name : hnd_mk
#rosbag files location
rosbag_location : /media/vassilis/Thesis/Datasets/PersonalFarm/TrainingActions
#pathname of calibrated image eges
cal_edges_path: /home/vassilis/Thesis/KinectPainting/Calibration_Edges.jpg
cal_frame_path: /home/vassilis/Thesis/KinectPainting/Calibration_Frame.jpg
#files to save images
save_depth: Depth
save_color: Color
#folders to save datasets and ground truths
test_save_path : /media/vassilis/Thesis/Datasets/PersonalFarm/actions/whole_result
ground_truth_fold: /media/vassilis/Thesis/Datasets/PersonalFarm/TrainingActions/ground_truths
#folder to save data
save_path: /media/vassilis/Thesis/Data/
class_save_path: /media/vassilis/Thesis/Data/Classifiers
#folder to display data
results_fold : /media/vassilis/Data/Thesis/KinectPainting/Results
#save parameters
max_depth_im_num_to_save: 200
# Choose object detection method ( segmentation or noise_model or mog2)
detection_method: mog2
# Variables necessary for object detection by using background noise model
max_contours_num: 3
calib_secs: 1
lowest_pixel_noise_deviation: 0.02
depth_thres: 10
lap_thres: 0.01
running_mean_depth_count: 5
outlier_shape_time: 1.5
#Variables for object detection by scene segmentation
similar_bg_min_dist: 2
cond_erosions_num: 10
segmentation_data: SegmData
max_objects_size_ratio: 0.04
min_displacement: 2
depth_tolerance: 20 #Too high, needs fixing
#Variables for object detection by mog2
gmm_num : 10
bg_ratio : 0.2
var_thres : 20
history : 4000
#Variables necessary for sparse coding
sparse_dim_rat: 2
sparse_alpha : 1
sparse_fss_max_iter: 1000
sparse_fss_min_coeff_rat: 0.05
sparse_fss_min_coeff: 3
sparse_fss_gamma: 1
#the rosbag for training
train_bag_path: /media/vassilis/Thesis/Datasets/PersonalFarm/TrainingActions/train.bag
#Variables for action recognition
ZHOF_bin_size: 4
3DHOF_bin_size: 4
GHOG_bin_size: 64
3DXYPCA_size: 32 # per dimension
GHOG_resize_size: 30 # in y dimension, proportional for x dimension
ZHOF_resize_size: 30 # in y dimension, proportional for x dimension
RDF_trees: 29
SVM_C: 83
AdaBoost_Estimators: 100
#Dynamic Params
STD_small_filt_window: 20 #window of buffers, not frames
STD_big_filt_window: 40 #window of buffers, not frames
action_separation_thres: 0
buffer_max_misses: 4
buffer_size: 10
buffer_confidence_tol: 0.25
filt_window_confidence_tol: 0.75
kernel : linear #for SVMs classifier
#PTPCA Params
PTPCA_components: 3
#Variables for hand segmentation
max_link_number : 6
max_hsa_contours_num: 5
min_area: 50
angle_resolution: 0.01
abnormality_tol: 1
angle_tol: 0.01
dist_tol: 1.1
longest_ray_closing_size: 15
HOG_bin_size : 30
HOG_max_window_length : 40
HOG_overlap : 20
#Variables to keep memory
memory_fade_const: 0.05
memory_fade_exp: 0.2
memory_power: 0.5
memory_strength: 0.2
# Variables for camshift
max_count: 80
par_eps: 300
hist_resol: 60
#Variables necessary for approximate wrist detection
interp_window: 7
cutoff_angle_ratio: 15
wearing_dist_rate: 0.05
lambda_power_weight: 1
length_ratio_power_weight: 1
length_power_weight: 1.5
num_of_checked_segments: 10
lambda_bound: 0.3
#Variables for data mining
free_ram_path: /media/vassilis/Thesis
#Variables for feature extraction
min_frame_count_diff: 5
max_gamma: 0.0001
#Show results and where (display,file,no)
results: display
#App configuration
AppData: AppData