请问大佬，我想用Ground Truth单独测试追踪的评价指标应该怎么做，就是我想用groundtruth作为追踪输入，得到追踪的结果，不用yolo跑出来的检测结果作为追踪的输入，该怎么做？我的ground truth是下面的格式
1,0,824,771,127,129,1,4,0,0
2,0,828,761,124,123,1,4,0,0
3,0,833,752,120,116,1,4,0,0
4,0,834,741,119,117,1,4,0,0
5,0,836,731,117,117,1,4,0,0
6,0,840,720,114,113,1,4,0,0
7,0,844,710,111,108,1,4,0,0
8,0,848,700,109,104,1,4,0,0
9,0,850,692,107,102,1,4,0,0
10,0,852,685,105,100,1,4,0,0
11,0,855,678,102,98,1,4,0,0
12,0,857,671,100,95,1,4,0,0
13,0,859,664,98,93,1,4,0,0
14,0,862,656,95,92,1,4,0,0
15,0,864,649,93,90,1,4,0,0
16,0,866,642,91,87,1,4,0,0
17,0,869,635,88,85,1,4,0,0
18,0,871,628,86,83,1,4,0,0
19,0,874,621,84,81,1,4,0,0
20,0,875,615,83,79,1,4,0,0
21,0,877,610,81,77,1,4,0,0
22,0,878,604,80,76,1,4,0,0
23,0,880,599,79,74,1,4,0,0

when I run the video_deepsort , an error is thrown :
File "C:/Users/16950/Desktop/yolo_deepsort-master/yolo_deepsort-master/video_deepsort.py", line 14, in
model.load_darknet_weights("weights/yolov3.weights")
File "C:\Users\16950\Desktop\yolo_deepsort-master\yolo_deepsort-master\yolo3\models\models.py", line 364, in load_darknet_weights
conv_w = torch.from_numpy(weights[ptr: ptr + num_w]).view_as(conv_layer.weight)
RuntimeError: shape '[1024, 512, 3, 3]' is invalid for input of size 2552319

I don't quite understand. Could you direct me？Thank you !

OSError: Couldn't open webcam or video

运行的时候 帧率很低 只有8帧

Hi @GlassyWing ,

i've managed to use your git project (working fine, nice work! :-) ) but I've got the problem that only ClassID 0 is tracked (I've got 142 trained classes).

I'm using my custom yolov4 weights file and the corresponding config.

Problem a.)
When i set classes=142 in my cfg, it throws an error:
File "/usr/local/src/yolo_deepsort/yolo3/models/models.py", line 199, in forward
prediction = x.view(num_samples, self.num_anchors, self.num_classes + 5, *grid_size).permute(0, 1, 3, 4, 2)
RuntimeError: shape '[1, 3, 147, 108, 108]' is invalid for input of size 2974320

Problem b is solved, so please ignore b) :-)

Can you give me some input how i can achieve that with your code or am i blind somewhere?

Thanks & Cheers! :-)

Problem b.) - [ Solved ] -> I had to reduce '...nms_thres=0.4....' down to '...nms_thres=0.1....' "

So when i use classes=80 from the coco default cfg its working fine. But my main problem is, that i want track all my classes not only classID 0. I've already set class_mask in video_deepsort.py but still no success (same problem also when i use the default yolov4.weights and the config for the coco dataset):

video_detector = VideoDetector(model,
....
class_mask=[0, 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, 5 6, 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]
....

Hi! Thank you for your work. I ran this repo before on google colab to track objects on a random video and it worked perfectly fine. However, I am failing to locate the tracker to the input video file right now. I'm not sure if I am doing anything wrong. I get the following error when I run video_deepsort.py:

Can you please help me out here?

When I run a test with a webcam or file, an error is thrown:
Traceback (most recent call last): File "video_deepsort.py", line 47, in <module> for image, detections, _ in video_detector.detect("rtsp://admin:[email protected]/ISAPI/Streaming/Channels/101", File "A:\ml\yolo_deepsort\yolo3\detect\video_detect.py", line 135, in detect detections = self.image_detector.detect(frame) File "A:\ml\yolo_deepsort\yolo3\detect\img_detect.py", line 86, in detect detections = self.model(image) File "C:\Users\pcdreams\miniconda3\envs\yolo_deepsort\lib\site-packages\torch\nn\modules\module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "A:\ml\yolo_deepsort\yolo3\models\models.py", line 308, in forward x, layer_loss = module[0](x, targets, img_dim) File "C:\Users\pcdreams\miniconda3\envs\yolo_deepsort\lib\site-packages\torch\nn\modules\module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "A:\ml\yolo_deepsort\yolo3\models\models.py", line 217, in forward pred_conf.view((num_samples, -1, 1)), RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.

whats a problem with view?

pytorch - 1.7.0
torchvision - 0.8.1
opencv-python - 4.4.0.46

Hi! I am running the repo on google colab. I have successfully able to use the code and run the tracking on my video but the output video is not saving even though I have specified the path.

According to the path, the output video is supposed to be stored at "data/" but after finishing the tracking the does not get saved. Can you please tell me how can I save the output prediction?

Thank you!

使用 python video_deepsort.py VIDEO_PATH为什么还是提示上面的错误？代码里还需要改什么吗

I will be appreciate the author if he/she can share the original traind model, thank you in advance.

你好,我在用您代码的时候检测不到是怎么回事?检测框不显示直接,希望您看到后回复我一下,谢谢

您好，我尝试yolov3-tiny,yolov4,yolov4-tiny都出现同样的问题，请问能否指点下，错误具体如下：
Traceback (most recent call last):
File "video_deepsort.py", line 50, in
skip_secs=0):
File "/root/yolo_deepsort/yolo3/detect/video_detect.py", line 141, in detect
detections = self.image_detector.detect(frame)
File "/root/yolo_deepsort/yolo3/detect/img_detect.py", line 85, in detect
detections = self.model(image)
File "/root/anaconda3/envs/python377/lib/python3.7/site-packages/torch/nn/modules/module.py", line 550, in call
result = self.forward(*input, **kwargs)
File "/root/yolo_deepsort/yolo3/models/models.py", line 299, in forward
x = module(x)
File "/root/anaconda3/envs/python377/lib/python3.7/site-packages/torch/nn/modules/module.py", line 550, in call
result = self.forward(*input, **kwargs)
File "/root/anaconda3/envs/python377/lib/python3.7/site-packages/torch/nn/modules/container.py", line 100, in forward
input = module(input)
File "/root/anaconda3/envs/python377/lib/python3.7/site-packages/torch/nn/modules/module.py", line 550, in call
result = self.forward(*input, **kwargs)
File "/root/anaconda3/envs/python377/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 349, in forward
return self._conv_forward(input, self.weight)
File "/root/anaconda3/envs/python377/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 346, in _conv_forward
self.padding, self.dilation, self.groups)
RuntimeError: "unfolded2d_copy" not implemented for 'Half'
terminate called without an active exception
Aborted (core dumped)

请问这个代码支持yolo4吗？

你好，请教下大佬，我想改进deepsort中reid模型，原论文是在market1501上训练的模型，对人的重识别效果好，对车的可能不好，所以我想在车的数据上训练出个模型，请问reid部分你和原论文做了哪些改动，还是保持没变，我该怎么训练？

请问运行test.py时出现了FileNotFoundError: [Errno 2] No such file or directory: 'data/coco/5k.txt'这样的错误是什么原因呢？

请问实时显示的视频中类别和编号都不清晰，有改进的空间吗？

if frames % self.skip_frames == 0:這麼處理的作用是什麼？skip_frames作用是什麼？
frames的值只有1和2，怎麼正常顯示幀數，並且能來一幀圖像處理一幀？

Hi,
1- How is the speed of Yolov3+DeepSort and Yolov4+DeepSort?
2- What are your system configurations?
3- Have you tested in Nvidia Jetson devices as well or not?

OSError: Couldn't open webcam or video

Hi I have been trying to run the repo in google colab with gpu. When I execute "video_deepsort.py" I get the following error:

2020-07-24 06:02:19,815 - INFO - Reading config...
2020-07-24 06:02:20,276 - INFO - Reading config done
Loading weights from weights/ckpt.t7... Done!
2020-07-24 06:02:24,310 - INFO - Detect video: data/test.mp4
2020-07-24 06:02:24,355 - INFO - Output Type: ../data/output.mp4, 828601953, 30, (1920, 1080)
2020-07-24 06:02:24,533 - INFO - Inference time: 0:00:00.072129
2020-07-24 06:02:24,616 - INFO - Inference time: 0:00:00.046835
2020-07-24 06:02:24,697 - INFO - Inference time: 0:00:00.041138
2020-07-24 06:02:24,762 - INFO - Inference time: 0:00:00.032771
2020-07-24 06:02:24,864 - INFO - Inference time: 0:00:00.057584
2020-07-24 06:02:24,939 - INFO - Inference time: 0:00:00.037254
2020-07-24 06:02:24,998 - INFO - Inference time: 0:00:00.030447
2020-07-24 06:02:25,090 - INFO - Inference time: 0:00:00.057867
2020-07-24 06:02:25,161 - INFO - Inference time: 0:00:00.046764
2020-07-24 06:02:25,231 - INFO - Inference time: 0:00:00.032764
2020-07-24 06:02:25,283 - INFO - Inference time: 0:00:00.029539
2020-07-24 06:02:25,370 - INFO - Inference time: 0:00:00.061455
FPS: 23
2020-07-24 06:02:25,468 - INFO - Inference time: 0:00:00.050386
2020-07-24 06:02:25,580 - INFO - Inference time: 0:00:00.055735
2020-07-24 06:02:25,647 - INFO - Inference time: 0:00:00.038648
2020-07-24 06:02:25,708 - INFO - Inference time: 0:00:00.030106
2020-07-24 06:02:25,763 - INFO - Inference time: 0:00:00.030028
2020-07-24 06:02:25,840 - INFO - Inference time: 0:00:00.049671
2020-07-24 06:02:25,898 - INFO - Inference time: 0:00:00.030984
2020-07-24 06:02:25,965 - INFO - Inference time: 0:00:00.037194
2020-07-24 06:02:26,085 - INFO - Inference time: 0:00:00.046189
2020-07-24 06:02:26,170 - INFO - Inference time: 0:00:00.048173
2020-07-24 06:02:26,237 - INFO - Inference time: 0:00:00.031856
2020-07-24 06:02:26,296 - INFO - Inference time: 0:00:00.030107
2020-07-24 06:02:26,383 - INFO - Inference time: 0:00:00.063005
FPS: 26
2020-07-24 06:02:26,465 - INFO - Inference time: 0:00:00.042469
2020-07-24 06:02:26,521 - INFO - Inference time: 0:00:00.029568
2020-07-24 06:02:26,578 - INFO - Inference time: 0:00:00.030526
2020-07-24 06:02:26,638 - INFO - Inference time: 0:00:00.030534
2020-07-24 06:02:26,703 - INFO - Inference time: 0:00:00.030067
2020-07-24 06:02:26,781 - INFO - Inference time: 0:00:00.049437
2020-07-24 06:02:26,877 - INFO - Inference time: 0:00:00.058676
2020-07-24 06:02:26,984 - INFO - Inference time: 0:00:00.037252
Exception in thread Thread-1:
Traceback (most recent call last):
File "/usr/lib/python3.6/threading.py", line 916, in _bootstrap_inner
self.run()
File "/usr/lib/python3.6/threading.py", line 864, in run
self._target(*self._args, **self._kwargs)
File "/usr/local/lib/python3.6/dist-packages/imutils/video/filevideostream.py", line 67, in update
frame = self.transform(frame)
File "/content/yolo_deepsort/yolo3/detect/video_detect.py", line 34, in _transform
return cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
cv2.error: OpenCV(4.1.2) /io/opencv/modules/imgproc/src/color.cpp:182: error: (-215:Assertion failed) !_src.empty() in function 'cvtColor'

2020-07-24 06:02:27,088 - INFO - Inference time: 0:00:00.036465
2020-07-24 06:02:27,138 - INFO - Inference time: 0:00:00.025053
2020-07-24 06:02:27,189 - INFO - Inference time: 0:00:00.024349
2020-07-24 06:02:27,238 - INFO - Inference time: 0:00:00.023730
2020-07-24 06:02:27,285 - INFO - Inference time: 0:00:00.024008
2020-07-24 06:02:27,332 - INFO - Inference time: 0:00:00.023859
2020-07-24 06:02:27,379 - INFO - Inference time: 0:00:00.023900
FPS: 30
2020-07-24 06:02:27,427 - INFO - Inference time: 0:00:00.023751
2020-07-24 06:02:27,476 - INFO - Inference time: 0:00:00.023747
2020-07-24 06:02:27,525 - INFO - Inference time: 0:00:00.024034
2020-07-24 06:02:27,578 - INFO - Inference time: 0:00:00.023974
2020-07-24 06:02:27,625 - INFO - Inference time: 0:00:00.023888
2020-07-24 06:02:27,675 - INFO - Inference time: 0:00:00.024328
2020-07-24 06:02:27,724 - INFO - Inference time: 0:00:00.023467
2020-07-24 06:02:27,773 - INFO - Inference time: 0:00:00.023828
2020-07-24 06:02:27,821 - INFO - Inference time: 0:00:00.023553
2020-07-24 06:02:27,876 - INFO - Inference time: 0:00:00.024036
2020-07-24 06:02:27,929 - INFO - Inference time: 0:00:00.023671
2020-07-24 06:02:27,986 - INFO - Inference time: 0:00:00.026806
2020-07-24 06:02:28,037 - INFO - Inference time: 0:00:00.023839
2020-07-24 06:02:28,092 - INFO - Inference time: 0:00:00.023777
2020-07-24 06:02:28,141 - INFO - Inference time: 0:00:00.024310
2020-07-24 06:02:28,190 - INFO - Inference time: 0:00:00.023869
2020-07-24 06:02:28,241 - INFO - Inference time: 0:00:00.024323
2020-07-24 06:02:28,293 - INFO - Inference time: 0:00:00.024464
2020-07-24 06:02:28,350 - INFO - Inference time: 0:00:00.026074
FPS: 38
2020-07-24 06:02:28,409 - INFO - Inference time: 0:00:00.027308
2020-07-24 06:02:28,465 - INFO - Inference time: 0:00:00.023894
2020-07-24 06:02:28,520 - INFO - Inference time: 0:00:00.023870
2020-07-24 06:02:28,579 - INFO - Inference time: 0:00:00.025258
2020-07-24 06:02:28,635 - INFO - Inference time: 0:00:00.023769
2020-07-24 06:02:28,689 - INFO - Inference time: 0:00:00.024628
2020-07-24 06:02:28,742 - INFO - Inference time: 0:00:00.024018
2020-07-24 06:02:28,795 - INFO - Inference time: 0:00:00.023710
2020-07-24 06:02:28,850 - INFO - Inference time: 0:00:00.024600
2020-07-24 06:02:28,905 - INFO - Inference time: 0:00:00.028612
2020-07-24 06:02:28,956 - INFO - Inference time: 0:00:00.023884
2020-07-24 06:02:29,009 - INFO - Inference time: 0:00:00.024233
2020-07-24 06:02:29,062 - INFO - Inference time: 0:00:00.023808
2020-07-24 06:02:29,117 - INFO - Inference time: 0:00:00.023849
2020-07-24 06:02:29,170 - INFO - Inference time: 0:00:00.023648
2020-07-24 06:02:29,224 - INFO - Inference time: 0:00:00.023403
2020-07-24 06:02:29,281 - INFO - Inference time: 0:00:00.024250
2020-07-24 06:02:29,338 - INFO - Inference time: 0:00:00.024134
FPS: 36
2020-07-24 06:02:29,392 - INFO - Inference time: 0:00:00.023779
2020-07-24 06:02:29,446 - INFO - Inference time: 0:00:00.023504
2020-07-24 06:02:29,501 - INFO - Inference time: 0:00:00.024275
2020-07-24 06:02:29,555 - INFO - Inference time: 0:00:00.023809
2020-07-24 06:02:29,608 - INFO - Inference time: 0:00:00.023458
2020-07-24 06:02:29,664 - INFO - Inference time: 0:00:00.024181
2020-07-24 06:02:29,720 - INFO - Inference time: 0:00:00.023965
2020-07-24 06:02:29,776 - INFO - Inference time: 0:00:00.025904
2020-07-24 06:02:29,836 - INFO - Inference time: 0:00:00.024041
2020-07-24 06:02:29,890 - INFO - Inference time: 0:00:00.023678
2020-07-24 06:02:29,952 - INFO - Inference time: 0:00:00.024716
2020-07-24 06:02:30,007 - INFO - Inference time: 0:00:00.024368
2020-07-24 06:02:30,062 - INFO - Inference time: 0:00:00.023714
2020-07-24 06:02:30,126 - INFO - Inference time: 0:00:00.023561

It looks like an opencv error to actually view the video while executing code. I am sure that colab won't allow to show the detection while running the code. I could be wrong but could you please suggest me something about this error?
Thanks in advance!