Skip to content

D
deepinfer
Commit 9461f84c authored by xin.wang.waytous's avatar xin.wang.waytous
Browse files
Options

all tasks done

parent 8b083a9d
Show whitespace changes
Inline Side-by-side
Showing with 1212 additions and 204 deletions
configs/tasks/bsw/bsw.yaml 0 → 100644
View file @ 9461f84c
configRootPath: /home/wangxin/projects/waytous/DeepInfer
device: 2
detectorName: CameraModel
detectorConfigPath: configs/tasks/bsw/rgb_yolov5.yaml
imageWidth: 1920
imageHeight: 1080
useFullPolygan: true
warningDistancePoolsPath: configs/tasks/bsw/left_camera_dist_v2.yaml
warningRangeConfigPath: configs/tasks/bsw/warning_infos.yaml
configs/tasks/bsw/left_camera_dist_v2.yaml 0 → 100644
View file @ 9461f84c
dist_100:
- - 7
- 992
- - 53
- 972
- - 102
- 959
- - 152
- 958
- - 204
- 961
- - 255
- 967
- - 305
- 972
- - 357
- 977
- - 407
- 982
- - 459
- 986
- - 512
- 990
- - 563
- 992
- - 615
- 996
- - 665
- 997
- - 715
- 999
- - 765
- 1000
- - 817
- 1001
- - 869
- 1001
- - 920
- 1002
- - 972
- 1002
- - 1024
- 1001
- - 1075
- 1000
- - 1126
- 999
- - 1176
- 997
- - 1227
- 995
- - 1280
- 992
- - 1331
- 989
- - 1382
- 986
- - 1432
- 982
- - 1482
- 980
- - 1533
- 983
- - 1583
- 994
- - 1633
- 1015
- - 1677
- 1041
- - 1717
- 1080
- - 0
- 1080
- - 0
- 992
dist_200:
- - 7
- 750
- - 56
- 734
- - 108
- 722
- - 158
- 717
- - 208
- 717
- - 260
- 718
- - 311
- 721
- - 361
- 723
- - 413
- 724
- - 463
- 726
- - 515
- 728
- - 565
- 729
- - 615
- 730
- - 667
- 731
- - 719
- 732
- - 769
- 733
- - 819
- 732
- - 869
- 733
- - 919
- 733
- - 970
- 733
- - 1023
- 733
- - 1076
- 732
- - 1129
- 731
- - 1181
- 731
- - 1233
- 730
- - 1284
- 728
- - 1334
- 727
- - 1384
- 725
- - 1436
- 726
- - 1487
- 729
- - 1538
- 733
- - 1588
- 742
- - 1639
- 759
- - 1689
- 776
- - 1737
- 798
- - 1783
- 824
- - 1827
- 855
- - 1866
- 888
- - 1903
- 925
- - 1920
- 925
- - 1920
- 1080
- - 0
- 1080
- - 0
- 750
dist_300:
- - 4
- 556
- - 53
- 539
- - 103
- 527
- - 153
- 518
- - 205
- 511
- - 256
- 509
- - 308
- 507
- - 358
- 507
- - 409
- 507
- - 460
- 507
- - 510
- 507
- - 560
- 507
- - 611
- 507
- - 663
- 507
- - 713
- 507
- - 763
- 507
- - 814
- 507
- - 865
- 507
- - 916
- 507
- - 967
- 507
- - 1018
- 507
- - 1069
- 507
- - 1120
- 507
- - 1170
- 507
- - 1220
- 507
- - 1272
- 507
- - 1323
- 507
- - 1374
- 509
- - 1424
- 511
- - 1476
- 514
- - 1527
- 522
- - 1579
- 532
- - 1628
- 543
- - 1676
- 558
- - 1726
- 576
- - 1775
- 597
- - 1820
- 619
- - 1863
- 645
- - 1905
- 673
- - 1920
- 673
- - 1920
- 1080
- - 0
- 1080
- - 0
- 556
dist_400:
- - 5
- 396
- - 53
- 380
- - 101
- 366
- - 152
- 355
- - 202
- 345
- - 252
- 338
- - 303
- 335
- - 354
- 331
- - 404
- 330
- - 454
- 329
- - 505
- 327
- - 555
- 326
- - 606
- 325
- - 656
- 324
- - 706
- 324
- - 756
- 324
- - 807
- 323
- - 858
- 323
- - 910
- 323
- - 961
- 323
- - 1012
- 323
- - 1063
- 323
- - 1114
- 324
- - 1164
- 324
- - 1214
- 325
- - 1264
- 326
- - 1315
- 327
- - 1366
- 329
- - 1416
- 334
- - 1466
- 338
- - 1519
- 348
- - 1570
- 357
- - 1619
- 369
- - 1669
- 385
- - 1718
- 400
- - 1766
- 419
- - 1814
- 440
- - 1859
- 462
- - 1904
- 488
- - 1920
- 488
- - 1920
- 1080
- - 0
- 1080
- - 0
- 396
dist_500:
- - 5
- 267
- - 54
- 251
- - 103
- 235
- - 153
- 222
- - 203
- 211
- - 254
- 202
- - 304
- 195
- - 354
- 190
- - 405
- 186
- - 456
- 184
- - 507
- 181
- - 557
- 179
- - 608
- 177
- - 659
- 177
- - 709
- 175
- - 759
- 174
- - 809
- 173
- - 860
- 174
- - 912
- 173
- - 962
- 173
- - 1012
- 173
- - 1062
- 174
- - 1113
- 175
- - 1164
- 176
- - 1216
- 177
- - 1267
- 179
- - 1318
- 182
- - 1368
- 185
- - 1419
- 191
- - 1470
- 198
- - 1521
- 207
- - 1571
- 218
- - 1620
- 230
- - 1670
- 245
- - 1718
- 261
- - 1766
- 280
- - 1814
- 299
- - 1860
- 322
- - 1906
- 346
- - 1920
- 346
- - 1920
- 1080
- - 0
- 1080
- - 0
- 267
configs/tasks/bsw/rgb_yolov5.yaml 0 → 100644
View file @ 9461f84c
name: CameraModel
inputNames: [cvImage]
outputNames: [trackedDetections]
units:
-
name: CameraSrc
inputNames: [cvImage]
outputNames: [uint8Image]
-
name: ResizeNorm
inputNames: [uint8Image]
outputNames: [resizeNormImages]
inputMean: [0, 0, 0]
inputStd: [1, 1, 1]
inferBatchSize: 1
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
useBGR: false
-
name: YoloV5TRTInference
inputNames: [resizeNormImages]
outputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes]
runMode: 1 # 0-fp32 1-fp16 2-int8 (int8 not supported)
weightsPath: "configs/tasks/bsw/yolov5s_rgb.wts"
engineFile: "configs/tasks/bsw/yolov5s_rgb_fp16.engine"
# calibImgPathFile: "configs/tasks/bsw/yolov5s_rgb_calib_imgs.txt"
# calibTableCache: "configs/tasks/bsw/yolov5s_rgb_calib_table.cache"
inferDynamic: true
inferBatchSize: 1
inputWidth: 640
inputHeight: 640
maxBatchSize: 4 # used when build engine
classNumber: 2
numAnchorsPerLevel: 3
numLevels: 3
startScale: 8
maxOutputBBoxCount: 1000
keepTopK: 100
# used when build engine, always be smaller than scoreThreshold.
# if you want it to take effect, you should delete the ${engineFile}
engineScoreThreshold: 0.1
nmsThreshold: 0.45
# yolov5
# n: 0.33+0.25 s: 0.33+0.5 m: 0.67+0.75 l: 1.0+1.0 x: 1.33+1.25
gd: 0.33
gw: 0.5
-
name: YoloV5PostProcess
inputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes, uint8Image]
outputNames: [detections]
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
scoreThreshold: 0.2 # used when inference, can be modified
truncatedThreshold: 0.05
keepTopK: 100
classNames: [vehicle, person]
-
name: ByteTracker
inputNames: [detections]
outputNames: [trackedDetections]
frame_rate: 20
track_buffer: 30
track_thresh: 0.2
high_thresh: 0.2
match_thresh: 0.8
init_frames: 0
configs/tasks/bsw/warning_infos.yaml 0 → 100644
View file @ 9461f84c
# 范围按照严重程度依次递增，即距离相机逐渐变近，默认情况为三个等级-green/yellow/red，green为无正常无警告
warningInfos:
-
name: Green
color: [0, 255, 0] # bgr format - green
warningFrameThreshold: 0
distance: -1 # m, -1 means inf
-
name: Yellow
color: [0, 255, 255] # bgr format - yellow
warningFrameThreshold: 3
distance: 2 # m
-
name: Red
color: [0, 0, 255] # bgr format - red
warningFrameThreshold: 3
distance: 1 # m
configs/tasks/dms/dms.yaml 0 → 100644
View file @ 9461f84c
configRootPath: /home/wangxin/projects/waytous/DeepInfer
device: 2
faceDetectorName: CameraModel
faceDetectorConfigPath: configs/tasks/dms/face_detect_yolov5.yaml
landmarkDetectorName: CameraModel
landmarkDetectorConfigPath: configs/tasks/dms/face_landmarks_mobilefacenet.yaml
headposeDetectorName: CameraModel
headposeDetectorConfigPath: configs/tasks/dms/face_headpose_whenet.yaml
mouthARThreshold: 0.75
eyeARThreshold: 0.25
eyeARConsecFrames: 15
distractionConsecFrames: 1
unmaskedConsecFrames: 20
nodriverConsecFrames: 20
sightYawThresholdLeft: -40
sightYawThresholdRight: 40
sightPitchThresholdLeft: -40
sightPitchThresholdRight: 40
configs/tasks/dms/face_detect_yolov5.yaml 0 → 100644
View file @ 9461f84c
name: CameraModel
inputNames: [cvImage]
outputNames: [detections]
units:
-
name: CameraSrc
inputNames: [cvImage]
outputNames: [uint8Image]
-
name: ResizeNorm
inputNames: [uint8Image]
outputNames: [resizeNormImages]
inputMean: [0, 0, 0]
inputStd: [1, 1, 1]
inferBatchSize: 1
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
useBGR: false
-
name: YoloV5TRTInference
inputNames: [resizeNormImages]
outputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes]
runMode: 1 # 0-fp32 1-fp16 2-int8 (int8 not supported)
weightsPath: "configs/tasks/dms/yolov5s_face.wts"
engineFile: "configs/tasks/dms/yolov5s_face_fp16.engine"
# calibImgPathFile: "configs/tasks/dms/yolov5s_calib_imgs.txt"
# calibTableCache: "configs/tasks/dms/yolov5s_calib_table.cache"
inferDynamic: true
inferBatchSize: 1
inputWidth: 640
inputHeight: 640
maxBatchSize: 1 # used when build engine
classNumber: 2
numAnchorsPerLevel: 3
numLevels: 3
startScale: 8
maxOutputBBoxCount: 1000
keepTopK: 100
# used when build engine, always be smaller than scoreThreshold.
# if you want it to take effect, you should delete the ${engineFile}
engineScoreThreshold: 0.1
nmsThreshold: 0.45
# yolov5
# n: 0.33+0.25 s: 0.33+0.5 m: 0.67+0.75 l: 1.0+1.0 x: 1.33+1.25
gd: 0.33
gw: 0.5
-
name: YoloV5PostProcess
inputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes, uint8Image]
outputNames: [detections]
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
scoreThreshold: 0.2 # used when inference, can be modified
truncatedThreshold: 0.05
keepTopK: 100
classNames: ["unmasked face", "masked face"]
configs/tasks/dms/face_headpose_whenet.yaml 0 → 100644
View file @ 9461f84c
name: CameraModel
inputNames: [cvFaceROIImage]
outputNames: [faceHeadpose]
units:
-
name: CameraSrc
inputNames: [cvFaceROIImage]
outputNames: [uint8FaceROIImage]
-
name: ResizeNorm
inputNames: [uint8FaceROIImage]
outputNames: [resizeFaceROIImage]
inputMean: [0.485, 0.456, 0.406]
inputStd: [0.229, 0.224, 0.225]
inferBatchSize: 1
inputWidth: 224
inputHeight: 224
fixAspectRatio: true
useBGR: false
-
name: TRTInference
inputNames: [resizeFaceROIImage]
outputNames: ["tf.identity", "tf.identity_1", "tf.identity_2"]
runMode: 1 # 0-fp32 1-fp16 2-int8 (int8 not supported)
weightsPath: "configs/tasks/dms/WHENet.onnx"
engineFile: "configs/tasks/dms/WHENet_fp16.engine"
# calibImgPathFile: "configs/tasks/dms/whenet_calib_imgs.txt"
# calibTableCache: "configs/tasks/dms/whenet_calib_table.cache"
inferDynamic: false
inferBatchSize: 1
inputWidth: 224
inputHeight: 224
maxBatchSize: 1 # used when build engine
-
name: WHENetPostProcess
inputNames: ["tf.identity", "tf.identity_1", "tf.identity_2"]
outputNames: [faceHeadpose]
outputYawLength: 120
outputPitchLength: 66
outputRollLength: 66
configs/tasks/dms/face_landmarks_mobilefacenet.yaml 0 → 100644
View file @ 9461f84c
name: CameraModel
inputNames: [cvFaceROIImage2]
outputNames: [faceLandmarks]
units:
-
name: CameraSrc
inputNames: [cvFaceROIImage2]
outputNames: [uint8FaceROIImage2]
-
name: ResizeNorm
inputNames: [uint8FaceROIImage2]
outputNames: [resizeFaceROIImage2]
inputMean: [0.0, 0.0, 0.0]
inputStd: [1.0, 1.0, 1.0]
inferBatchSize: 1
inputWidth: 112
inputHeight: 112
fixAspectRatio: true
useBGR: false
-
name: TRTInference
inputNames: [resizeFaceROIImage2]
outputNames: ["output.0"]
runMode: 1 # 0-fp32 1-fp16 2-int8 (int8 not supported)
weightsPath: "configs/tasks/dms/mobilefacenet_model.onnx"
engineFile: "configs/tasks/dms/mobilefacenet_model_fp16.engine"
# calibImgPathFile: "configs/tasks/dms/mobilefacenet_calib_imgs.txt"
# calibTableCache: "configs/tasks/dms/mobilefacenet_calib_table.cache"
inferDynamic: false
inferBatchSize: 1
inputWidth: 112
inputHeight: 112
maxBatchSize: 1 # used when build engine
-
name: MobileFaceNetPostProcess
inputNames: ["output.0", uint8FaceROIImage2]
outputNames: [faceLandmarks]
inputWidth: 112
inputHeight: 112
fixAspectRatio: true
landmarkNumber: 68
configs/tasks/fusion/fusion.yaml 0 → 100644
View file @ 9461f84c
configRootPath: /home/wangxin/projects/waytous/DeepInfer
device: 2
modelName: CameraModel
modelConfigPath: configs/tasks/fusion/fusion_model.yaml
# undistort image for vis
mainSrcName: undistortVisImage
configs/tasks/fusion/fusion_model.yaml 0 → 100644
View file @ 9461f84c
name: CameraModel
inputNames: [cvVisImage, cvThermalImage]
# outputNames: [fusionedDetections]
outputNames: [trackedDetections]
units:
# camera rgb, using TraDes segmentor
-
name: CameraSrc
inputNames: [cvVisImage]
outputNames: [uint8VisImage]
-
name: Undistort
inputNames: [uint8VisImage]
outputNames: [undistortVisImage]
imageWidth: 1920
imageHeight: 1080
IntrinsicPath: configs/tasks/fusion/rgb_intrinsic.yaml
-
name: ResizeNorm
inputNames: [undistortVisImage]
outputNames: [resizeVisImage]
inputMean: [0.40789654, 0.44719302, 0.47026115]
inputStd: [0.28863828, 0.27408164, 0.27809835]
useBGR: true
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
fixAspectRatio: false
-
name: TRTInference
inputNames: [resizeVisImage]
outputNames: [hm, reg, wh, conv_weight, seg_feat]
runMode: 1 # 0-fp32 1-fp16 2-int8 (int8 not supported)
weightsPath: "configs/tasks/mots/trades_segv3.onnx"
engineFile: "configs/tasks/mots/trades_segv3_fp16.engine"
# calibImgPathFile: "configs/tasks/mots/trades_calib_imgs.txt"
# calibTableCache: "configs/tasks/mots/trades_calib_table.cache"
inferDynamic: false
maxBatchSize: 1
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
-
name: TraDesPostProcess
inputNames: [hm, reg, wh, conv_weight, seg_feat, uint8VisImage]
outputNames: [visDetections]
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
fixAspectRatio: false
scoreThreshold: 0.2
truncatedThreshold: 0.05
classNumber: 3
classNames: [person, vehicle, cycle]
topK: 100
downScale: 4
segDims: 64
maxCntsLength: 500
# thermal with yolov5
-
name: CameraSrc
inputNames: [cvThermalImage]
outputNames: [uint8ThermalImage]
-
name: Undistort
inputNames: [uint8ThermalImage]
outputNames: [undistortThermalImage]
imageWidth: 640
imageHeight: 512
IntrinsicPath: configs/tasks/fusion/thermal_intrinsic.yaml
-
name: ResizeNorm
inputNames: [undistortThermalImage]
outputNames: [resizeThermalImage]
inputMean: [0, 0, 0]
inputStd: [1, 1, 1]
useBGR: false
inferBatchSize: 1
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
-
name: YoloV5TRTInference
inputNames: [resizeThermalImage]
outputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes]
runMode: 1 # 0-fp32 1-fp16 2-int8 (int8 not supported)
weightsPath: "configs/tasks/thermal/yolov5s_640x640_thermal.wts"
engineFile: "configs/tasks/thermal/yolov5s_640x640_thermal_fp16.engine"
# calibImgPathFile: "configs/tasks/thermal/yolov5s_calib_imgs.txt"
# calibTableCache: "configs/tasks/thermal/yolov5s_calib_table.cache"
inferDynamic: true
inferBatchSize: 1
inputWidth: 640
inputHeight: 640
maxBatchSize: 4 # using when build engine
classNumber: 2
numAnchorsPerLevel: 3
numLevels: 3
startScale: 8
maxOutputBBoxCount: 1000
keepTopK: 100
# used when build engine, always be smaller than scoreThreshold.
# if you want it to take effect, you should delete the ${engineFile}
engineScoreThreshold: 0.1
nmsThreshold: 0.45
# yolov5
# n: 0.33+0.25 s: 0.33+0.5 m: 0.67+0.75 l: 1.0+1.0 x: 1.33+1.25
gd: 0.33
gw: 0.5
-
name: YoloV5PostProcess
inputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes, uint8ThermalImage]
outputNames: [thermalDetections]
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
scoreThreshold: 0.2 # used when inference, can be modified
truncatedThreshold: 0.05
keepTopK: 100
classNames: [vehicle, person]
-
name: HomoProject
inputNames: [thermalDetections]
outputNames: [thermalProjectDetections]
homoMatrixPath: configs/tasks/fusion/thermal_homography.yaml
# fusion
-
name: BayesianFusioner
inputNames: [visDetections, thermalProjectDetections]
outputNames: [fusionedDetections]
NMSThreshold: 0.5
# supported: IOU IOA GIOU CIOU
matchMethod: IOA
# tracker
-
name: ByteTracker
inputNames: [fusionedDetections]
outputNames: [trackedDetections]
frame_rate: 20
track_buffer: 30
track_thresh: 0.2
high_thresh: 0.2
match_thresh: 0.8
init_frames: 0
configs/tasks/fusion/rgb_intrinsic.yaml 0 → 100644
View file @ 9461f84c
image_width: 1920
image_height: 1080
camera_name: narrow_stereo/left
camera_matrix:
rows: 3
cols: 3
data: [ 954.94556, 0. , 959.48891,
0. , 955.72757, 550.42017,
0. , 0. , 1. ]
camera_model: plumb_bob
distortion_coefficients:
rows: 1
cols: 5
data: [-0.308855, 0.080618, -0.000948, -0.000307, 0.000000]
rectification_matrix:
rows: 3
cols: 3
data: [ 0.99964588, -0.02369215, -0.01211637,
0.02381908, 0.99966175, 0.01044166,
0.01186489, -0.01072657, 0.99987207]
projection_matrix:
rows: 3
cols: 4
data: [ 863.43207, 0. , 964.79819, 0. ,
0. , 863.43207, 544.12591, 0. ,
0. , 0. , 1. , 0. ]
configs/tasks/fusion/thermal_homography.yaml 0 → 100644
View file @ 9461f84c
dst_height: 1080
dst_width: 1920
homography_matrix:
rows: 3
cols: 3
data: [ 1.43434450e+00, -2.82390527e-02, 4.89822323e+02,
1.96702059e-02, 1.29985905e+00, 1.87569822e+02,
1.51906922e-04, -1.06563084e-04, 1.00000000e+00
]
configs/tasks/fusion/thermal_intrinsic.yaml 0 → 100644
View file @ 9461f84c
image_width: 640
image_height: 512
camera_name: thermal
camera_matrix:
rows: 3
cols: 3
data: [ 755.0212, 2.0979 , 311.1998,
0. , 757.7410, 261.7544,
0. , 0. , 1. ]
camera_model: plumb_bob
distortion_coefficients:
rows: 1
cols: 5
data: [-0.4137, 0.1847, -0.0056, 0.0014, 0.0186]
# rectification_matrix:
# rows: 3
# cols: 3
# data:
# projection_matrix:
# rows: 3
# cols: 4
# data:
configs/tasks/mots/trades.yaml
View file @ 9461f84c
name: TraDeS
name: CameraModel
inputNames: [cvImage1]
outputNames: [trackedDetections1]
globalParams:
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
fixAspectRatio: false
units:
-
name: CameraSrc
......@@ -18,6 +14,10 @@ units:
outputNames: [resizeNormImages]
inputMean: [0.40789654, 0.44719302, 0.47026115]
inputStd: [0.28863828, 0.27408164, 0.27809835]
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
fixAspectRatio: false
useBGR: true
-
name: TRTInference
......@@ -30,14 +30,21 @@ units:
# calibTableCache: "configs/tasks/mots/trades_calib_table.cache"
inferDynamic: false
maxBatchSize: 1
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
-
name: TraDesPostProcess
inputNames: [hm, reg, wh, conv_weight, seg_feat, uint8Image1]
outputNames: [segInstances1]
inferBatchSize: 1
inputWidth: 640
inputHeight: 384
fixAspectRatio: false
scoreThreshold: 0.2
truncatedThreshold: 0.05
classNumber: 3
classNames: [vehicle, person, cycle]
classNames: [person, vehicle, cycle]
topK: 100
downScale: 4
segDims: 64
......
configs/tasks/thermal/yolov5.yaml
View file @ 9461f84c
......@@ -2,11 +2,7 @@
name: CameraModel
inputNames: [cvImage1, cvImage2]
outputNames: [trackedDetections1, trackedDetections2]
globalParams:
inferBatchSize: 2
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
units:
-
name: CameraSrc
......@@ -22,6 +18,10 @@ units:
outputNames: [resizeNormImages]
inputMean: [0, 0, 0]
inputStd: [1, 1, 1]
inferBatchSize: 2
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
useBGR: false
-
name: YoloV5TRTInference
......@@ -33,7 +33,10 @@ units:
# calibImgPathFile: "configs/tasks/thermal/yolov5s_calib_imgs.txt"
# calibTableCache: "configs/tasks/thermal/yolov5s_calib_table.cache"
inferDynamic: true
maxBatchSize: 4
inferBatchSize: 2
inputWidth: 640
inputHeight: 640
maxBatchSize: 4 # used when build engine
classNumber: 2
numAnchorsPerLevel: 3
numLevels: 3
......@@ -52,6 +55,9 @@ units:
name: YoloV5PostProcess
inputNames: [num_detections, nmsed_boxes, nmsed_scores, nmsed_classes, uint8Image1, uint8Image2]
outputNames: [detections1, detections2]
inputWidth: 640
inputHeight: 640
fixAspectRatio: true
scoreThreshold: 0.2 # used when inference, can be modified
truncatedThreshold: 0.05
keepTopK: 100
......
main.cpp
View file @ 9461f84c
......@@ -37,7 +37,10 @@ int main(int argc, char** argv){
std::string srcPath = argv[4];
std::string savePath = argv[5];
auto t = interfaces::BaseTaskRegisterer::GetInstanceByName(taskName);
t->Init(configPath);
if(!t->Init(configPath)){
printf("Init problems!\n");
return 0;
};
printf("Init all Done\n");
auto paths = stringSplit(srcPath, ',');
......@@ -64,11 +67,17 @@ int main(int argc, char** argv){
auto e2 = std::chrono::system_clock::now();
std::cout << "100 times avg infer time: " <<
std::chrono::duration_cast<std::chrono::microseconds>(e2 - e1).count() / 1000. / 100. << " ms" << std::endl;
if(inputs.size() != outputs.size()){
cv::Mat vis = images[0];
t->Visualize(&vis, outputs[0]);
cv::imwrite(savePaths[0], vis);
}else{
for(int i=0; i<inputs.size(); i++){
cv::Mat vis = images[i];
t->Visualize(&vis, outputs[i]);
cv::imwrite(savePaths[i], vis);
}
}
}else if(videoOrImage == "video"){
cv::VideoCapture cap(srcPath);
......@@ -98,7 +107,20 @@ int main(int argc, char** argv){
/*
./main TaskMOTS image ../configs/tasks/mots/mots.yaml ../test/bsw.jpg ../test/mots_res.jpg
./main TaskDetect image ../configs/tasks/thermal/thermal.yaml ../test/thermal2.jpg,../test/thermal2.jpg ../test/t_b1_res.jpg,../test/t_b2_res.jpg
./main TaskFusion image ../configs/tasks/fusion/fusion.yaml ../test/fusion_rgb.png,../test/fusion_thermal.png ../test/fusion_res.jpg
./main TaskDMS image ../configs/tasks/dms/dms.yaml ../test/dms.jpg ../test/dms_res.jpg
./main TaskBSW image ../configs/tasks/bsw/bsw.yaml ../test/bsw.jpg ../test/bsw_res.jpg
*/
......
src/base/image.h
View file @ 9461f84c
......@@ -74,6 +74,9 @@ public:
offset_ = 0;
channels_ = image->channels();
blob_.reset(new Blob<uint8_t>({rows_, cols_, channels_}));
if(image->data == nullptr){
LOG_ERROR << "image data null";
}
blob_->set_cpu_data(image->data);
width_step_ = blob_->offset({1, 0, 0}) * static_cast<int>(sizeof(uint8_t));
}
......
src/common/common.h
View file @ 9461f84c
......@@ -60,7 +60,7 @@ inline float euclidean(float x1, float y1, float x2, float y2){
}
void softmax(float *x, int length){
inline void softmax(float *x, int length){
float max = x[0];
float sum = 0.0;
for(int i=0; i<length; i++){
......
src/interfaces/base_model.h
View file @ 9461f84c
......@@ -35,6 +35,7 @@ public:
}
public:
std::vector<BaseUnitPtr> units_;
std::vector<std::string> inputNames;
std::vector<std::string> outputNames;
......
src/interfaces/base_unit.cpp
View file @ 9461f84c
......@@ -4,13 +4,9 @@ namespace waytous {
namespace deepinfer {
namespace interfaces {
bool BaseUnit::Init(YAML::Node& node, YAML::Node& globalParamNode){
bool BaseUnit::Init(YAML::Node& node){
inputNames = node["inputNames"].as<std::vector<std::string>>();
outputNames = node["outputNames"].as<std::vector<std::string>>();
inferBatchSize = globalParamNode["inferBatchSize"].as<int>(); // default=1
inputHeight = globalParamNode["inputHeight"].as<int>();
inputWidth = globalParamNode["inputWidth"].as<int>();
fixAspectRatio = globalParamNode["fixAspectRatio"].as<bool>();
return true;
};
......
src/interfaces/base_unit.h
View file @ 9461f84c
......@@ -22,7 +22,7 @@ public:
BaseUnit() = default;
virtual ~BaseUnit() = default;
virtual bool Init(YAML::Node& node, YAML::Node& globalParamNode);
virtual bool Init(YAML::Node& node);
virtual bool Exec() = 0;
......@@ -34,10 +34,6 @@ public:
std::vector<std::string> inputNames;
std::vector<std::string> outputNames;
//global params
int inferBatchSize = 1;
int inputWidth, inputHeight;
bool fixAspectRatio = true;
};
using BaseUnitPtr = std::shared_ptr<BaseUnit>;
......
src/libs/fusioner/bayesian_fusioner.cpp
View file @ 9461f84c
......@@ -5,16 +5,16 @@ namespace deepinfer {
namespace fusioner {
bool BayesianFusioner::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
LOG_WARN << "Init trades postprocess error";
bool BayesianFusioner::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init BayesianFusioner error";
return false;
};
NMSThreshold = node["NMSThreshold"].as<float>();
std::string methodStr = node["matchMethod"].as<std::string>();
auto method = MethodName2Type.find(methodStr);
if(method == MethodName2Type.end()){
LOG_WARN << "Not supported match method " << methodStr;
LOG_WARN << "BayesianFusioner not supported match method " << methodStr;
return false;
}
matchMethod = method->second;
......@@ -47,6 +47,7 @@ bool BayesianFusioner::Exec(){
);
auto fusioned_objects = std::make_shared<ios::Detection2Ds>(ios::Detection2Ds());
interfaces::SetIOPtr(outputNames[0], fusioned_objects);
for(size_t i=0; i < indices.size(); i++){
if(!deleted[indices[i]]){
std::vector<ios::Det2DPtr> matched_objs;
......@@ -60,15 +61,21 @@ bool BayesianFusioner::Exec(){
matched_objs.push_back(all_objects[indices[j]]);
}
}
/* merge score, useless
float pos_sum = 0;
float neg_sum = 0;
for(auto& obj : matched_objs){
// merge matched mask
if(main_obj->mask_ptr == nullptr){
main_obj->mask_ptr = obj->mask_ptr;
}
pos_sum += std::log(obj->confidence);
neg_sum += std::log(1 - obj->confidence);
}
pos_sum = std::exp(pos_sum);
neg_sum = std::exp(neg_sum);
main_obj->confidence = pos_sum / (pos_sum + neg_sum);
*/
// if()
fusioned_objects->detections.push_back(main_obj);
}
......
src/libs/fusioner/bayesian_fusioner.h
View file @ 9461f84c
......@@ -31,7 +31,7 @@ public:
{"IOA", BayesianFusionMatchMethod::IOA}
};
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
......
src/libs/inferences/tensorrt/trt_infer.cpp
View file @ 9461f84c
......@@ -22,12 +22,15 @@ TRTInference::~TRTInference(){
}
bool TRTInference::Init(YAML::Node& configNode, YAML::Node& globalParamNode){
if(!BaseUnit::Init(configNode, globalParamNode)){
bool TRTInference::Init(YAML::Node& configNode){
if(!BaseUnit::Init(configNode)){
LOG_WARN << "Init trt engine error";
return false;
};
inputHeight = configNode["inputHeight"].as<int>();
inputWidth = configNode["inputWidth"].as<int>();
inferBatchSize = configNode["inferBatchSize"].as<int>();
engineFile = configNode["engineFile"].as<std::string>();
inferDynamic = configNode["inferDynamic"].as<bool>();
engineFile = common::GetAbsolutePath(common::ConfigRoot::GetRootPath(), engineFile);
......
src/libs/inferences/tensorrt/trt_infer.h
View file @ 9461f84c
......@@ -32,7 +32,7 @@ class TRTInference: public interfaces::BaseUnit {
public:
~TRTInference();
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
virtual bool BuildEngine(YAML::Node& node);
bool Exec() override;
virtual std::string Name() override;
......@@ -51,6 +51,8 @@ public:
std::string engineFile;
bool inferDynamic = false;
int inferBatchSize = 1;
int inputWidth, inputHeight;
}; // class TRTInferenceEngine
DEEPINFER_REGISTER_UNIT(TRTInference);
......
src/libs/postprocessors/homo_project.cpp
View file @ 9461f84c
......@@ -5,26 +5,29 @@ namespace deepinfer {
namespace postprocess {
bool HomoProject::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
bool HomoProject::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init homo project error";
return false;
};
auto homoMatrixPath = node["homoMatrixPath"].as<std::string>();
std::string yaml_path = common::GetAbsolutePath(common::ConfigRoot::GetRootPath(), homoMatrixPath);
if (!common::PathExists(yaml_path)) {
LOG_INFO << yaml_path << " does not exist!";
LOG_INFO << "Load homo project matrix " << yaml_path << " does not exist!";
return false;
}
YAML::Node node = YAML::LoadFile(yaml_path);
if (node.IsNull()) {
LOG_INFO << "Load " << yaml_path << " failed! please check!";
YAML::Node homoNode = YAML::LoadFile(yaml_path);
if (homoNode.IsNull()) {
LOG_INFO << "Load homo project matrix " << yaml_path << " failed! please check!";
return false;
}
dst_width = homoNode["dst_width"].as<int>();
dst_height = homoNode["dst_height"].as<int>();
for(int i=0; i<9; i++){
camera_homo_to_main(i) = node["homography_matrix"]["data"][i].as<float>();
camera_homo_to_main(i) = homoNode["homography_matrix"]["data"][i].as<float>();
}
return true;
}
......@@ -54,6 +57,15 @@ bool HomoProject::Exec(){
projected_obj->y1 = pbox(2);
projected_obj->y2 = pbox(3);
projected_obj->image_height = dst_height;
projected_obj->image_width = dst_width;
projected_obj->validCoordinate();
// todo mask_project
if(projected_obj->mask_ptr != nullptr){
}
projected_dets->detections.push_back(projected_obj);
}
interfaces::SetIOPtr(outputNames[0], projected_dets);
......
src/libs/postprocessors/homo_project.h
View file @ 9461f84c
......@@ -17,13 +17,14 @@ namespace postprocess {
class HomoProject: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
std::string Name() override;
public:
int dst_height, dst_width;
Eigen::Matrix<float, 3, 3, Eigen::RowMajor> camera_homo_to_main;
};
......
src/libs/postprocessors/mobilefacenet_post.cpp
View file @ 9461f84c
......@@ -5,11 +5,14 @@ namespace deepinfer {
namespace postprocess {
bool MobileFaceNetPostProcess::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
LOG_WARN << "Init trades postprocess error";
bool MobileFaceNetPostProcess::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init mobilefacenet postprocess error";
return false;
};
inputHeight = node["inputHeight"].as<int>();
inputWidth = node["inputWidth"].as<int>();
fixAspectRatio = node["fixAspectRatio"].as<bool>();
landmarkNumber = node["landmarkNumber"].as<int>();
return true;
}
......@@ -17,13 +20,13 @@ bool MobileFaceNetPostProcess::Init(YAML::Node& node, YAML::Node& globalParamNod
bool MobileFaceNetPostProcess::Exec(){
if (inputNames.size() != 1 + outputNames.size()){
LOG_ERROR << "yolov5 postprocess, inputsize != 4 + ouputsize.";
LOG_ERROR << "mobilefacenet postprocess, inputsize != 1 + ouputsize.";
return false;
}
auto input = std::dynamic_pointer_cast<ios::NormalIO>(interfaces::GetIOPtr(inputNames[0]));
if (input == nullptr){
LOG_ERROR << "WHENet postprocess input " << inputNames[0] << " haven't been init or doesn't exist.";
LOG_ERROR << "mobilefacenet postprocess input " << inputNames[0] << " haven't been init or doesn't exist.";
return false;
}
......@@ -32,7 +35,7 @@ bool MobileFaceNetPostProcess::Exec(){
auto iName = inputNames[j];
auto iptr = std::dynamic_pointer_cast<ios::CameraSrcOut>(interfaces::GetIOPtr(iName));
if (iptr == nullptr){
LOG_ERROR << "YoloV5 postprocess input " << iName << " haven't been init or doesn't exist.";
LOG_ERROR << "mobilefacenet postprocess input " << iName << " haven't been init or doesn't exist.";
return false;
}
inputImages.push_back(iptr->img_ptr_);
......
src/libs/postprocessors/mobilefacenet_post.h
View file @ 9461f84c
......@@ -20,13 +20,15 @@ namespace postprocess {
class MobileFaceNetPostProcess: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
std::string Name() override;
public:
int inputHeight, inputWidth;
bool fixAspectRatio = true;
int landmarkNumber;
};
......
src/libs/postprocessors/trades_post.cpp
View file @ 9461f84c
......@@ -5,12 +5,16 @@ namespace deepinfer {
namespace postprocess {
bool TraDesPostProcess::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
bool TraDesPostProcess::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init trades postprocess error";
return false;
};
inputHeight = node["inputHeight"].as<int>();
inputWidth = node["inputWidth"].as<int>();
inferBatchSize = node["inferBatchSize"].as<int>();
fixAspectRatio = node["fixAspectRatio"].as<bool>();
scoreThreshold = node["scoreThreshold"].as<float>();
truncatedThreshold = node["truncatedThreshold"].as<float>();
classNumber = node["classNumber"].as<int>();
......
src/libs/postprocessors/trades_post.h
View file @ 9461f84c
......@@ -21,7 +21,7 @@ namespace postprocess {
class TraDesPostProcess: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
......@@ -34,6 +34,9 @@ public:
base::BlobPtr<int> maskCnts_lengths_ptr;
base::BlobPtr<int> maskCnts_ptr;
int inputHeight, inputWidth;
int inferBatchSize = 1;
bool fixAspectRatio = true;
float scoreThreshold, truncatedThreshold;
int classNumber;
std::vector<std::string> classNames;
......
src/libs/postprocessors/whenet_post.cpp
View file @ 9461f84c
......@@ -5,9 +5,9 @@ namespace deepinfer {
namespace postprocess {
bool WHENetPostProcess::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
LOG_WARN << "Init trades postprocess error";
bool WHENetPostProcess::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init WHENet postprocess error";
return false;
};
outputYawLength = node["outputYawLength"].as<int>();
......@@ -34,36 +34,38 @@ bool WHENetPostProcess::Exec(){
float* roll = inputs[1]->mutable_cpu_data();
float* pitch = inputs[2]->mutable_cpu_data();
for(int b=0; b<outputNames.size(); b++){
auto headpose = std::make_shared<ios::HeadPose>(ios::HeadPose());
interfaces::SetIOPtr(outputNames[0], headpose);
interfaces::SetIOPtr(outputNames[b], headpose);
float yawSum = 0, pitchSum=0, rollSum = 0;
// yaw
common::softmax(yaw, outputYawLength);
common::softmax(yaw + b * outputYawLength, outputYawLength);
for(int i=0; i<outputYawLength; i++){
yawSum += yaw[i] * i;
yawSum += yaw[i + b * outputYawLength] * i;
}
yawSum *= 3.0;
yawSum -= 180;
headpose->yaw = yawSum;
// roll
common::softmax(roll, outputRollLength);
common::softmax(roll + b * outputRollLength, outputRollLength);
for(int i=0; i<outputRollLength; i++){
rollSum += roll[i] * i;
rollSum += roll[i + b * outputRollLength] * i;
}
rollSum *= 3.0;
rollSum -= 99;
headpose->roll = rollSum;
// pitch
common::softmax(pitch, outputPitchLength);
common::softmax(pitch + b * outputPitchLength, outputPitchLength);
for(int i=0; i<outputPitchLength; i++){
pitchSum += pitch[i] * i;
pitchSum += pitch[i + b * outputPitchLength] * i;
}
pitchSum *= 3.0;
pitchSum -= 99;
headpose->pitch = pitchSum;
}
return true;
}
......
src/libs/postprocessors/whenet_post.h
View file @ 9461f84c
......@@ -20,7 +20,7 @@ namespace postprocess {
class WHENetPostProcess: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
......
src/libs/postprocessors/yolov5_post.cpp
View file @ 9461f84c
......@@ -5,11 +5,14 @@ namespace deepinfer {
namespace postprocess {
bool YoloV5PostProcess::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
LOG_WARN << "Init trades postprocess error";
bool YoloV5PostProcess::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init yolov5 postprocess error";
return false;
};
inputHeight = node["inputHeight"].as<int>();
inputWidth = node["inputWidth"].as<int>();
fixAspectRatio = node["fixAspectRatio"].as<bool>();
classNames = node["classNames"].as<std::vector<std::string>>();
truncatedThreshold = node["truncatedThreshold"].as<float>(); // default 5%
scoreThreshold = node["scoreThreshold"].as<float>();
......
src/libs/postprocessors/yolov5_post.h
View file @ 9461f84c
......@@ -20,13 +20,15 @@ namespace postprocess {
class YoloV5PostProcess: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
std::string Name() override;
public:
int inputHeight, inputWidth;
bool fixAspectRatio = true;
int keepTopK = 100;
float scoreThreshold, truncatedThreshold;
std::vector<std::string> classNames;
......
src/libs/preprocessors/resize_norm.cpp
View file @ 9461f84c
......@@ -6,11 +6,16 @@ namespace deepinfer {
namespace preprocess {
bool ResizeNorm::Init(YAML::Node& node, YAML::Node& globalParamNode){
if(!BaseUnit::Init(node, globalParamNode)){
bool ResizeNorm::Init(YAML::Node& node){
CUDA_CHECK(cudaStreamCreate(&stream_));
if(!BaseUnit::Init(node)){
LOG_WARN << "Init resize_norm error";
return false;
};
inputHeight = node["inputHeight"].as<int>();
inputWidth = node["inputWidth"].as<int>();
inferBatchSize = node["inferBatchSize"].as<int>();
fixAspectRatio = node["fixAspectRatio"].as<bool>();
if(inputNames.size() != inferBatchSize){
LOG_ERROR << "Resize norm got wrong inputs number: " << inputNames.size() << " with infer batchsize: " << inferBatchSize;
......
src/libs/preprocessors/resize_norm.h
View file @ 9461f84c
......@@ -21,16 +21,19 @@ namespace preprocess {
class ResizeNorm: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
std::string Name() override;
public:
cudaStream_t stream_;
base::BlobPtr<float> dst, mean, std;
int inputHeight, inputWidth;
int inferBatchSize = 1;
bool fixAspectRatio = true;
bool useBGR = false;
base::BlobPtr<float> dst, mean, std;
cudaStream_t stream_;
};
......
src/libs/preprocessors/undistort.cpp
View file @ 9461f84c
......@@ -5,8 +5,8 @@ namespace deepinfer {
namespace preprocess {
bool Undistort::Init(YAML::Node& node, YAML::Node& globalParamNode) {
if(!BaseUnit::Init(node, globalParamNode)){
bool Undistort::Init(YAML::Node& node) {
if(!BaseUnit::Init(node)){
LOG_WARN << "Init trades postprocess error";
return false;
};
......@@ -16,18 +16,10 @@ bool Undistort::Init(YAML::Node& node, YAML::Node& globalParamNode) {
std::string IntrinsicPath, ExtrinsicPath = "";
IntrinsicPath = node["IntrinsicPath"].as<std::string>();
if(node["ExtrinsicPath"].IsDefined() && !node["ExtrinsicPath"].IsNull()){
ExtrinsicPath = node["ExtrinsicPath"].as<std::string>();
}
if(!loadIntrinsic(IntrinsicPath)){
LOG_WARN << "Load intrinsic error: " << IntrinsicPath;
return false;
}
if(ExtrinsicPath != "" && !loadExtrinsics(ExtrinsicPath)){
LOG_WARN << "Load Extrinsic error: " << ExtrinsicPath;
return false;
}
d_mapx_.Reshape({height_, width_});
d_mapy_.Reshape({height_, width_});
......@@ -40,66 +32,14 @@ bool Undistort::Init(YAML::Node& node, YAML::Node& globalParamNode) {
camera_intrinsic, width_,
height_, &d_mapx_, &d_mapy_);
dst_img = std::make_shared<base::Image8U>(base::Image8U(height_, width_, base::Color::BGR));
auto output = std::make_shared<ios::CameraSrcOut>(ios::CameraSrcOut(dst_img));
interfaces::SetIOPtr(outputNames[0], output);
inited_ = true;
return true;
}
bool Undistort::loadExtrinsics(std::string& yaml_file){
std::string yaml_path = common::GetAbsolutePath(common::ConfigRoot::GetRootPath(), yaml_file);
if (!common::PathExists(yaml_path)) {
LOG_INFO << yaml_path << " does not exist!";
return false;
}
YAML::Node node = YAML::LoadFile(yaml_path);
double qw = 0.0;
double qx = 0.0;
double qy = 0.0;
double qz = 0.0;
double tx = 0.0;
double ty = 0.0;
double tz = 0.0;
try {
if (node.IsNull()) {
LOG_INFO << "Load " << yaml_path << " failed! please check!";
return false;
}
qw = node["transform"]["rotation"]["w"].as<double>();
qx = node["transform"]["rotation"]["x"].as<double>();
qy = node["transform"]["rotation"]["y"].as<double>();
qz = node["transform"]["rotation"]["z"].as<double>();
tx = node["transform"]["translation"]["x"].as<double>();
ty = node["transform"]["translation"]["y"].as<double>();
tz = node["transform"]["translation"]["z"].as<double>();
} catch (YAML::InvalidNode &in) {
LOG_ERROR << "load camera extrisic file " << yaml_path
<< " with error, YAML::InvalidNode exception";
return false;
} catch (YAML::TypedBadConversion<double> &bc) {
LOG_ERROR << "load camera extrisic file " << yaml_path
<< " with error, YAML::TypedBadConversion exception";
return false;
} catch (YAML::Exception &e) {
LOG_ERROR << "load camera extrisic file " << yaml_path
<< " with error, YAML exception:" << e.what();
return false;
}
camera_extrinsic.setConstant(0);
Eigen::Quaterniond q;
q.x() = qx;
q.y() = qy;
q.z() = qz;
q.w() = qw;
camera_extrinsic.block<3, 3>(0, 0) = q.normalized().toRotationMatrix();
camera_extrinsic(0, 3) = tx;
camera_extrinsic(1, 3) = ty;
camera_extrinsic(2, 3) = tz;
camera_extrinsic(3, 3) = 1;
return true;
}
bool Undistort::loadIntrinsic(std::string& yaml_file){
std::string yaml_path = common::GetAbsolutePath(common::ConfigRoot::GetRootPath(), yaml_file);
if (!common::PathExists(yaml_path)) {
......@@ -211,8 +151,6 @@ bool Undistort::Exec(){
}
auto src_img = iptr->img_ptr_;
auto dst_img = std::make_shared<base::Image8U>(base::Image8U(src_img->rows(), src_img->cols(), src_img->type()));
auto output = std::make_shared<ios::CameraSrcOut>(ios::CameraSrcOut(dst_img));
NppiInterpolationMode remap_mode = NPPI_INTER_LINEAR;
NppiSize image_size;
......@@ -247,7 +185,6 @@ bool Undistort::Exec(){
return false;
}
interfaces::SetIOPtr(outputNames[0], output);
return true;
}
......
src/libs/preprocessors/undistort.h
View file @ 9461f84c
......@@ -22,14 +22,13 @@ namespace preprocess {
class Undistort: public interfaces::BaseUnit{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
std::string Name() override;
bool loadIntrinsic(std::string& yaml_file);
bool loadExtrinsics(std::string& yaml_file);
void InitUndistortRectifyMap(const Eigen::Matrix3f &camera_model,
const Eigen::Matrix<float, 1, 5> &distortion,
const Eigen::Matrix3f &R,
......@@ -40,7 +39,7 @@ public:
public:
base::Blob<float> d_mapx_;
base::Blob<float> d_mapy_;
Eigen::Matrix<double, 4, 4, Eigen::RowMajor> camera_extrinsic; // load by another yaml file
base::Image8UPtr dst_img;
Eigen::Matrix<float, 3, 3, Eigen::RowMajor> camera_intrinsic;
Eigen::Matrix<float, 3, 3, Eigen::RowMajor> camera_rectification;
Eigen::Matrix<float, 3, 4, Eigen::RowMajor> camera_projection;
......@@ -48,9 +47,6 @@ public:
int width_ = 0; // image cols
int height_ = 0; // image rows
int in_size_ = 0; // size of the input image in byte
int out_size_ = 0; // size of the output image in byte
int device_ = 0; // device number for gpu
bool inited_ = 0;
};
......
src/libs/trackers/byte_tracker.cpp
View file @ 9461f84c
......@@ -7,10 +7,10 @@ namespace tracker{
bool ByteTracker::Init(YAML::Node& node, YAML::Node& globalParamNode)
bool ByteTracker::Init(YAML::Node& node)
{
if(!BaseUnit::Init(node, globalParamNode)){
LOG_WARN << "Init resize_norm error";
if(!BaseUnit::Init(node)){
LOG_WARN << "Init ByteTracker error";
return false;
};
int frame_rate = node["frame_rate"].as<int>();
......@@ -22,18 +22,17 @@ bool ByteTracker::Init(YAML::Node& node, YAML::Node& globalParamNode)
frame_id = 0;
max_time_lost = int(frame_rate / 30.0 * track_buffer);
LOG_INFO << "Init ByteTracker!";
return true;
}
bool ByteTracker::Exec()
{
if(interfaces::GetIOPtr(inputNames[0]) == nullptr){
LOG_ERROR << "ByteTracker input" << inputNames[0] << " haven't init";
auto input = std::dynamic_pointer_cast<ios::Detection2Ds>(interfaces::GetIOPtr(inputNames[0]));
if(input == nullptr){
LOG_ERROR << "ByteTracker input " << inputNames[0] << " haven't init";
return false;
}
auto input = std::dynamic_pointer_cast<ios::Detection2Ds>(interfaces::GetIOPtr(inputNames[0]));
auto objects = input->detections;
////////////////// Step 1: Get detections //////////////////
......
src/libs/trackers/byte_tracker.h
View file @ 9461f84c
......@@ -18,7 +18,7 @@ class ByteTracker: public interfaces::BaseUnit
{
public:
bool Init(YAML::Node& node, YAML::Node& globalParamNode) override;
bool Init(YAML::Node& node) override;
bool Exec() override;
std::string Name() override;
......
src/models/camera_model.cpp
View file @ 9461f84c
......@@ -9,28 +9,27 @@ bool CameraModel::Init(std::string& configPath) {
// CUDA_CHECK(cudaStreamCreate(&stream_));
std::string cfgPath = common::GetAbsolutePath(common::ConfigRoot::GetRootPath(), configPath);
if(!common::PathExists(cfgPath)){
LOG_WARN << "config_file "<< common::ConfigRoot::GetRootPath() << " " << cfgPath << " not exist.";
LOG_WARN << "Init CameraModel config_file "<< common::ConfigRoot::GetRootPath() << " " << cfgPath << " not exist.";
return false;
}
modelConfigNode = YAML::LoadFile(cfgPath);
if (modelConfigNode.IsNull()) {
LOG_WARN << "Load " << configPath << " failed! please check!";
LOG_WARN << "Init CameraModel, Load " << configPath << " failed! please check!";
return false;
}
inputNames = modelConfigNode["inputNames"].as<std::vector<std::string>>();
outputNames = modelConfigNode["outputNames"].as<std::vector<std::string>>();
auto globalParamNode = modelConfigNode["globalParams"];
// units
auto unitNodes = modelConfigNode["units"];
for(int i=0; i<unitNodes.size(); i++){
auto unitNode = unitNodes[i];
std::string unitName = unitNode["name"].as<std::string>();
LOG_INFO << "Init unit: " << unitName;
LOG_INFO << "Init CameraModel unit: " << unitName;
interfaces::BaseUnitPtr unitPtr;
unitPtr.reset(interfaces::BaseUnitRegisterer::GetInstanceByName(unitName));
if(!unitPtr->Init(unitNode, globalParamNode)){
LOG_WARN << "Init unit " << unitName << " failed!";
if(!unitPtr->Init(unitNode)){
LOG_WARN << "Init CameraModel unit " << unitName << " failed!";
return false;
};
units_.push_back(unitPtr);
......@@ -44,22 +43,22 @@ bool CameraModel::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::Bas
for (int i=0; i<inputs.size(); i++){
auto input_img = inputs[i];
auto inputName = inputNames[i];
LOG_INFO << "input: " << inputName;
LOG_INFO << "CameraModel input: " << inputName;
auto input = std::make_shared<ios::CameraSrcIn>(ios::CameraSrcIn(input_img));
interfaces::SetIOPtr(inputName, input);
}
// infer
for(auto unit: units_){
LOG_INFO << "exec unit: " << unit->Name();
for(auto& unit: units_){
LOG_INFO << "exec unit " << unit->Name();
if(!unit->Exec()){
LOG_ERROR << unit->Name() << " unit exec error";
LOG_ERROR << "exec CameraModel unit "<< unit->Name() << " exec error";
return false;
};
}
// output
for(auto outName: outputNames){
LOG_INFO << "output: " << outName;
LOG_INFO << "CameraModel output: " << outName;
outputs.push_back(interfaces::GetIOPtr(outName));
}
......@@ -75,9 +74,14 @@ bool CameraModel::Exec(std::vector<base::Image8UPtr> inputs, std::vector<interfa
auto input = std::make_shared<ios::CameraSrcOut>(ios::CameraSrcOut(input_img));
interfaces::SetIOPtr(inputName, input);
}
// infer
for(auto unit: units_){
unit->Exec();
for(auto& unit: units_){
LOG_INFO << "exec unit" << unit->Name();
if(!unit->Exec()){
LOG_ERROR << "exec CameraModel unit "<< unit->Name() << " exec error";
return false;
};
}
// output
......
src/models/camera_model.h
View file @ 9461f84c
......@@ -25,7 +25,6 @@ public:
public:
YAML::Node modelConfigNode;
std::vector<interfaces::BaseUnitPtr> units_;
};
DEEPINFER_REGISTER_MODELS(CameraModel);
......
src/tasks/task_bsw.cpp
View file @ 9461f84c
......@@ -8,7 +8,7 @@ namespace task{
bool TaskBSW::Init(std::string& taskConfigPath){
if(!interfaces::BaseTask::Init(taskConfigPath)){
LOG_ERROR << "Init task detect error";
LOG_ERROR << "Init task BSW error";
return false;
};
std::string detectorName = taskNode["detectorName"].as<std::string>();
......@@ -16,7 +16,7 @@ bool TaskBSW::Init(std::string& taskConfigPath){
detector.reset(interfaces::BaseModelRegisterer::GetInstanceByName(detectorName));
if(!detector->Init(detectorConfigPath)){
LOG_ERROR << detectorName << " detector init problem";
LOG_ERROR << "Init BSW "<< detectorName << " detector init problem";
return false;
};
......@@ -26,14 +26,15 @@ bool TaskBSW::Init(std::string& taskConfigPath){
// init distance pools
std::string warningDistancePoolsPath = taskNode["warningDistancePoolsPath"].as<std::string>();
useFullPolygan = taskNode["useFullPolygan"].as<bool>();
std::string distanceConfigPath = common::GetAbsolutePath(common::ConfigRoot::GetRootPath(), warningDistancePoolsPath);
if(!common::PathExists(distanceConfigPath)){
LOG_ERROR << "camera distance pool config_file " << distanceConfigPath << " not exist.";
LOG_ERROR << "Init BSW, camera distance pool config_file " << distanceConfigPath << " not exist.";
return false;
}
YAML::Node distanceNode = YAML::LoadFile(distanceConfigPath);
if (distanceNode.IsNull()) {
LOG_ERROR << "Load " << distanceConfigPath << " failed! please check!";
LOG_ERROR << "Init BSW, Load" << distanceConfigPath << " failed! please check!";
return false;
}
......@@ -49,12 +50,12 @@ bool TaskBSW::Init(std::string& taskConfigPath){
bool TaskBSW::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOPtr>& outputs){
if(inputs.size() != 1){
LOG_ERROR << 'Now only support infer one image once.';
LOG_ERROR << "Task BSW now only support infer one image once.";
return false;
}
std::vector<interfaces::BaseIOPtr> dets;
if(!detector->Exec(inputs, dets)){
LOG_ERROR << "Task Detect Exec error";
LOG_ERROR << "Task BSW's detector exec error";
return false;
};
......@@ -112,15 +113,16 @@ bool TaskBSW::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOP
void TaskBSW::Visualize(cv::Mat* image, interfaces::BaseIOPtr outs){
for(auto &bo_pair : BSWRes){
auto& bo = bo_pair.second;
auto bo = bo_pair.second;
if(bo->lost_ages > 0){
continue;
}
auto &obj = bo->obj;
cv::Scalar color = bo->counters[0] % 4 == 0 ? cv::Scalar(255, 255, 255) : ranges[(int) bo->msg].color;
cv::rectangle(*image, cv::Rect(int(obj->x1), int(obj->y1), int(obj->x2 - obj->x1), int(obj->x2 - obj->y1)), color, 2);
auto obj = bo->obj;
cv::Scalar color = ranges[(int) bo->msg].color; //bo->counters[0] % 4 == 0 ? cv::Scalar(255, 255, 255) :
cv::rectangle(*image, cv::Rect(int(obj->x1), int(obj->y1), int(obj->x2 - obj->x1), int(obj->y2 - obj->y1)), color, 2);
cv::putText(*image, std::to_string(obj->track_id) + "." + obj->class_name + ":" + common::formatValue(obj->confidence, 2),
cv::Point(int(obj->x1), int(obj->y1) - 5), cv::FONT_HERSHEY_SIMPLEX, 0.5, color, 1);
}
// draw outlines
cv::addWeighted(*image, 0.9, mask, 0.1, 0, *image);
......@@ -141,7 +143,6 @@ bool TaskBSW::InitRanges(const std::string& warningRangeConfigPath, YAML::Node&
LOG_ERROR << "Load " << configPath << " failed! please check!";
return false;
}
bool useFullPolygan = taskNode["useFullPolygan"].as<bool>();
auto infoNode = rangeNode["warningInfos"];
for(int i=0; i < infoNode.size(); i++){
RangeInfo rinfo;
......
src/tasks/task_bsw.h
View file @ 9461f84c
......@@ -63,6 +63,7 @@ public:
std::vector<RangeInfo> ranges;
std::map<int, BSWResultPtr> BSWRes;
int imageWidth, imageHeight;
bool useFullPolygan = true;
cv::Mat mask;
};
......
src/tasks/task_dms.cpp
View file @ 9461f84c
......@@ -8,7 +8,7 @@ namespace task{
bool TaskDMS::Init(std::string& taskConfigPath){
if(!interfaces::BaseTask::Init(taskConfigPath)){
LOG_ERROR << "Init task detect error";
LOG_ERROR << "Init task dms error";
return false;
};
std::string faceDetectorName = taskNode["faceDetectorName"].as<std::string>();
......@@ -31,17 +31,17 @@ bool TaskDMS::Init(std::string& taskConfigPath){
faceDetector.reset(interfaces::BaseModelRegisterer::GetInstanceByName(faceDetectorName));
if(!faceDetector->Init(faceDetectorConfigPath)){
LOG_ERROR << faceDetectorName << " faceDetector init problem";
LOG_ERROR << "Init task DMS " << faceDetectorName << " faceDetector init problem";
return false;
};
landmarkDetector.reset(interfaces::BaseModelRegisterer::GetInstanceByName(landmarkDetectorName));
if(!landmarkDetector->Init(landmarkDetectorConfigPath)){
LOG_ERROR << landmarkDetectorName << " face landmarks detector init problem";
LOG_ERROR << "Init task DMS " << landmarkDetectorName << " face landmarks detector init problem";
return false;
};
headposeDetector.reset(interfaces::BaseModelRegisterer::GetInstanceByName(headposeDetectorName));
if(!headposeDetector->Init(headposeDetectorConfigPath)){
LOG_ERROR << headposeDetectorName << " face headpose detector init problem";
LOG_ERROR << "Init task DMS " << headposeDetectorName << " face headpose detector init problem";
return false;
};
......@@ -52,7 +52,7 @@ bool TaskDMS::Init(std::string& taskConfigPath){
bool TaskDMS::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOPtr>& outputs){
if(inputs.size() != 1){
LOG_ERROR << 'Now only support infer one image once.';
LOG_ERROR << "DMS now only support infer one image once.";
return false;
}
std::vector<interfaces::BaseIOPtr> faceDets;
......@@ -68,7 +68,7 @@ bool TaskDMS::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOP
NODRIVER_COUNTER++;
if(NODRIVER_COUNTER >= nodriverConsecFrames){
res->msg = DMSMsg::DMS_NODRIVER;
LOG_INFO << "No driver!";
// LOG_INFO << "No driver!";
}else
res->msg = DMSMsg::DMS_NONE;
return true;
......@@ -108,13 +108,13 @@ bool TaskDMS::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOP
// set landmarks and whenet input src
std::vector<cv::Mat*> subInputs;
auto faceImage = (*inputs[0])(faceROI);
auto faceImage = (*inputs[0])(faceROI).clone(); // Mat(Rect) -> temporary reference, need clone() to copy data
subInputs.push_back(&faceImage);
// face landmarks
std::vector<interfaces::BaseIOPtr> landmarks_vec;
if(!landmarkDetector->Exec(subInputs, landmarks_vec)){
LOG_ERROR << "face landmarks detect doesnot end normally!";
LOG_ERROR << "Task DMS face landmarks detect doesnot end normally!";
return false;
};
res->face_landmarks = std::dynamic_pointer_cast<ios::Landmarks>(landmarks_vec[0]);
......@@ -141,7 +141,7 @@ bool TaskDMS::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOP
// headpose
std::vector<interfaces::BaseIOPtr> headpose_vec;
if(!headposeDetector->Exec(subInputs, headpose_vec)){
LOG_ERROR << "face headpose detect doesnot end normally!";
LOG_ERROR << "Task DMS face headpose detect doesnot end normally!";
return false;
};
res->headPose = std::dynamic_pointer_cast<ios::HeadPose>(headpose_vec[0]);
......
src/tasks/task_fusion.cpp
View file @ 9461f84c
......@@ -8,15 +8,16 @@ namespace task{
bool TaskFusion::Init(std::string& taskConfigPath){
if(!interfaces::BaseTask::Init(taskConfigPath)){
LOG_ERROR << "Init task detect error";
LOG_ERROR << "Init task fusion error";
return false;
};
std::string detectorName = taskNode["detectorName"].as<std::string>();
std::string detectorConfigPath = taskNode["detectorConfigPath"].as<std::string>();
mainSrcName = taskNode["mainSrcName"].as<std::string>();
std::string modelName = taskNode["modelName"].as<std::string>();
std::string modelConfigPath = taskNode["modelConfigPath"].as<std::string>();
model.reset(interfaces::BaseModelRegisterer::GetInstanceByName(detectorName));
if(!model->Init(detectorConfigPath)){
LOG_ERROR << detectorName << " detector init problem";
model.reset(interfaces::BaseModelRegisterer::GetInstanceByName(modelName));
if(!model->Init(modelConfigPath)){
LOG_ERROR << "Init Fusion "<< modelName << " model init problem";
return false;
};
return true;
......@@ -25,7 +26,7 @@ bool TaskFusion::Init(std::string& taskConfigPath){
bool TaskFusion::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOPtr>& outputs){
if(!model->Exec(inputs, outputs)){
LOG_ERROR << "Task Detect Exec error";
LOG_ERROR << "Task fusion Exec error";
return false;
};
return true;
......@@ -33,15 +34,49 @@ bool TaskFusion::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::Base
void TaskFusion::Visualize(cv::Mat* image, interfaces::BaseIOPtr outs){
auto mainSrc = std::dynamic_pointer_cast<ios::CameraSrcOut>(interfaces::GetIOPtr(mainSrcName));
auto mainImage = mainSrc->img_ptr_->toCVMat();
auto detections = std::dynamic_pointer_cast<ios::Detection2Ds>(outs)->detections;
for(auto &obj : detections){
// LOG_INFO << "box: " << obj->x1 << " " << obj->x2 << " " << obj->y1 << " " << obj->y2;
cv::Scalar color = get_color(obj->class_label);
cv::rectangle(*image, cv::Rect(int(obj->x1), int(obj->y1), int(obj->x2 - obj->x1), int(obj->y2 - obj->y1)), color, 2);
cv::putText(*image, common::formatValue(obj->confidence, 2) + ":" +
std::to_string(obj->track_id) + "." + obj->class_name,
cv::Point(int(obj->x1), int(obj->y1) - 5), cv::FONT_HERSHEY_SIMPLEX, 0.5, color, 1);
cv::Mat realInstanceMask, colorInstance;
for(auto& obj: detections){
cv::Scalar color = get_color(obj->class_label * 100 + obj->track_id);
cv::putText(mainImage, std::to_string(obj->track_id) + ":" + common::formatValue(obj->confidence, 2),
cv::Point(obj->x1, obj->y1 - 5),
0, 0.6, cv::Scalar(0, 0, 255), 2, cv::LINE_AA);
cv::rectangle(mainImage, cv::Rect(obj->x1, obj->y1,
obj->x2 - obj->x1, obj->y2 - obj->y1), color, 2);
if(obj->mask_ptr != nullptr){
cv::Mat instance_mask = obj->mask_ptr->decode();
instance_mask.convertTo(instance_mask, CV_32FC1);
cv::resize(instance_mask, realInstanceMask, mainImage.size());
mainImage.copyTo(colorInstance);
for(int i=0; i<realInstanceMask.rows; i++){
for(int j=0; j<realInstanceMask.cols; j++){
if(realInstanceMask.at<float>(i, j) >= 0.5){
colorInstance.at<cv::Vec3b>(i, j) = cv::Vec3b(color[0], color[1], color[2]);
}
}
}
cv::addWeighted(mainImage, 0.5, colorInstance, 0.5, 0, mainImage);
}
}
/*for(int i=0; i<cameraHandlers.size(); i++){
if(i != mainCameraIndex){
cv::Mat img = cameraHandlers[i].frame->image_ptr->toCVMat();
cv::Mat dst;
if(cameraHandlers[i].frame->camera_info_ptr->sensor_type == base::SensorType::THERMAL_CAMERA){
cv::applyColorMap(img, img, cv::COLORMAP_JET);
}
cv::Mat homoMatrix;
cv::eigen2cv(cameraHandlers[i].frame->camera_info_ptr->camera_homo_to_main, homoMatrix);
cv::warpPerspective(img, dst, homoMatrix, mainImage.size());
mainImage = cv::max(mainImage, dst);
}
}*/
image->data = mainImage.data;
}
......
src/tasks/task_fusion.h
View file @ 9461f84c
......@@ -2,6 +2,9 @@
#ifndef WAYTOUS_DEEPINFER_TASK_FUSION_H_
#define WAYTOUS_DEEPINFER_TASK_FUSION_H_
#include <Eigen/Core>
#include <opencv2/core/eigen.hpp>
#include "interfaces/base_task.h"
#include "models/camera_model.h"
#include "libs/ios/detection.h"
......@@ -22,6 +25,7 @@ public:
public:
std::shared_ptr<interfaces::BaseModel> model;
std::string mainSrcName;
};
......
src/tasks/task_mots.cpp
View file @ 9461f84c
......@@ -8,7 +8,7 @@ namespace task{
bool TaskMOTS::Init(std::string& taskConfigPath){
if(!interfaces::BaseTask::Init(taskConfigPath)){
LOG_ERROR << "Init task detect error";
LOG_ERROR << "Init task MOTS error";
return false;
};
std::string segmentorName = taskNode["segmentorName"].as<std::string>();
......@@ -16,7 +16,7 @@ bool TaskMOTS::Init(std::string& taskConfigPath){
segmentor.reset(interfaces::BaseModelRegisterer::GetInstanceByName(segmentorName));
if(!segmentor->Init(segmentorConfigPath)){
LOG_ERROR << segmentorName << " segmentor init problem";
LOG_ERROR << "Task MOTS "<< segmentorName << " segmentor init problem";
return false;
};
return true;
......@@ -25,7 +25,7 @@ bool TaskMOTS::Init(std::string& taskConfigPath){
bool TaskMOTS::Exec(std::vector<cv::Mat*> inputs, std::vector<interfaces::BaseIOPtr>& outputs){
if(!segmentor->Exec(inputs, outputs)){
LOG_ERROR << "Task Detect Exec error";
LOG_ERROR << "Task MOTS Exec error";
return false;
};
return true;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment