File size: 3,647 Bytes
c747942 | 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 | # Copyright 2026 Yokohama National University. All Rights Reserved.
# This code specifies the file paths for the input features and the unified functional spectrum label data used in the preprocessing scripts 1-1 and 1-2.
# session name
session = ['2024_0807session1F','2024_0807session3F',
'2024_0809session1F','2024_0809session3F',
'2024_0826session1F','2024_0826session3F',
'2024_0829session1F','2024_0829session3F',
'2024_0905session1F','2024_0905session3F',
'2024_0910session1F','2024_0910session3F']
# participant name
P = ['P1','P2','P3','P4']
# Pair name of speaker, addressee, and other listeners
# Speaker–addressee order is ignored (e.g., P1_P2_P3P4 == P2_P1_P3P4)
Pair = ['P1_P2_P3P4','P1_P3_P2P4','P1_P4_P2P3','P2_P3_P1P4','P2_P4_P1P3','P3_P4_P1P2']
# Input feature paths
head_pose_angle_ = ['data/head_pose_angle_data/','.openface.txt'] # Head pose angle path
utter_ = ['data/utter_data/','.utter.txt'] # Utterance activity path
gaze_status_GT_ = ['data/gaze_status_GT_data/','.gaze.status_GT.txt'] # Gaze state (ground truth) path
eyeball_ = ['data/eyeball_data/','.OpenFace.EyeforCNN.txt'] # Gaze direction path
au_ = ['data/au_data/','.OpenFace.corrFrame.AU_norm.','.txt'] # Facial expression AU data path
# Data paths for conditional branching
floor_holder_ = ['data/FloorHolder_data/','.FloorHolder.mainlabeler.txt'] # Floor holder path
gaze_direction_ = ['data/gaze_direction_data/','.gaze_label.withTransition.txt'] # Gaze direction path
# Unified Functional Spectrum label data paths
Unified_functional_spectrum_pair_ = ['data/Unified_functional_spectrum_pair/H_','.txt'] # Path for each participant pair
Unified_functional_spectrum_debate_ = ['data/Unified_functional_spectrum_debate/H_','.txt'] # Path for each debate (date)
ave_Unified_functional_spectrum_pair_ = ['data/Unified_functional_spectrum_pair/ave_','.txt'] # Path storing the average label values
head_pose_angle = {}
utter = {}
gaze_status_GT = {}
eyeball = {}
au = {}
floor_holder = {}
Unified_functional_spectrum_pair = {}
ave_Unified_functional_spectrum_pair = {}
Unified_functional_spectrum_debate = {}
gaze_direction = {}
# Detailed path settings for head pose angles, utterance status, gaze status predictions,
# eyeball directions, AU features, unified functional spectrum label data,
# and data used for conditional branching
for debate in session:
head_pose_angle[debate] = head_pose_angle_[0] + debate + head_pose_angle_[1]
utter[debate] = utter_[0] + debate + utter_[1]
gaze_status_GT[debate] = gaze_status_GT_[0] + debate + gaze_status_GT_[1]
eyeball[debate] = eyeball_[0] + debate + eyeball_[1]
floor_holder[debate] = floor_holder_[0] + debate + floor_holder_[1]
gaze_direction[debate] = gaze_direction_[0] + debate + gaze_direction_[1]
Unified_functional_spectrum_debate[debate] = Unified_functional_spectrum_debate_[0] + debate + Unified_functional_spectrum_debate_[1]
for person in P:
au[debate + person] = au_[0] + debate + au_[1] + person + au_[2]
for pair in Pair:
Unified_functional_spectrum_pair[debate + pair] = Unified_functional_spectrum_pair_[0] + debate + pair + Unified_functional_spectrum_pair_[1]
ave_Unified_functional_spectrum_pair[debate + pair] = ave_Unified_functional_spectrum_pair_[0] + debate + pair + ave_Unified_functional_spectrum_pair_[1]
# Save path
save = 'processed_data/'
save_TF = ['false/','true/'] |