运行ope_otb，加载数据库OTB10时，卡死在14%，您知道可能的原因吗？
loading OTB100: 13%|█████▏ | 13/100
[01:04<18loading OTB100: 13%|█████ | 13/100
[01:05<18:loading OTB100: 14%|█████▍ | 14/100
[01:19<19:loading OTB100: 14%|█████▋ | 14/100 [01:19<19:22, 13.51s/it, Box]
电脑运行到这里卡死！！！！！！！

there is no data to even run an example. your description in readme is incorrent. python cf_demo.py not working. and in repo no folder dataset, and also no any documentation about format or content of this directory. wtf?

PS C:\Users\username\Desktop\pyCFTrackers\examples> python .\cf_demo.py Traceback (most recent call last): File ".\cf_demo.py", line 7, in <module> data_names=sorted(os.listdir(data_dir)) FileNotFoundError: [WinError 3] Системе не удается найти указанный путь: '../dataset/test' PS C:\Users\username\Desktop\pyCFTrackers\examples>

who is running this code without gpu？How it worked in cpu system

Hello,
Thank you for sharing.
I try BACF on my server and only ~10fps (OTB100), w/o loading images. Could u tell me the speed of the BACF when u test ?

Thanks for your help!

Hello,
I have test the BACF with the toolkit GOT-10k (https://github.com/got-10k/toolkit) on OTB100, but the score is much lower than the pysot. (I can't run the pysot on my server and have no idea.)

GOT10k: | precision 0.622 | sucess 0.462
pysot: | precision 0.796 | sucess 0.598

Do you have any idea about this?
Is there any other toolkit can evaluate the python tracker?

Thank you for any reply.

I’m sorry to disturb u with this question. I have checked the issues in pyCFTrackers and pysot-toolkit, but have no answers.

I use the eval_OTB.py to test OTB100

It raise errors：

File "pysot/utils/statistics.py", line 104:
def success_overlap(gt_bb, result_bb, n_frame):
thresholds_overlap = np.arange(0, 1.05, 0.05)
^

This error may have been caused by the following argument(s):

• argument 1: Unsupported array dtype: object

This is not usually a problem with Numba itself but instead often caused by
the use of unsupported features or an issue in resolving types.

To see Python/NumPy features supported by the latest release of Numba visit:
http://numba.pydata.org/numba-doc/dev/reference/pysupported.html
and
http://numba.pydata.org/numba-doc/dev/reference/numpysupported.html

For more information about typing errors and how to debug them visit:
http://numba.pydata.org/numba-doc/latest/user/troubleshoot.html#my-code-doesn-t-compile

If you think your code should work with Numba, please report the error message
and traceback, along with a minimal reproducer at:
https://github.com/numba/numba/issues/new

And when i comment the # @jit(nopython=True), it raise errors：

"""
Traceback (most recent call last):
File "/root/anaconda3/envs/cf/lib/python3.5/multiprocessing/pool.py", line 119, in worker
result = (True, func(*args, **kwds))
File "/root/Documents/pysot-toolkit-master/pysot/evaluation/ope_benchmark.py", line 50, in eval_success
success_ret_[video.name] = success_overlap(gt_traj, tracker_traj, n_frame)
File "/root/Documents/pysot-toolkit-master/pysot/utils/statistics.py", line 108, in success_overlap
iou[mask] = overlap_ratio(gt_bb[mask], result_bb[mask])
IndexError: too many indices for array
"""

Thank you for any reply! ! !

File "", line 1, in
runfile('/home/zly/pyCFTrackers/eval/eval_VOT2016.py', wdir='/home/zly/pyCFTrackers/eval')

File "/home/zly/anaconda3/lib/python3.7/site-packages/spyder_kernels/customize/spydercustomize.py", line 786, in runfile
execfile(filename, namespace)

File "/home/zly/anaconda3/lib/python3.7/site-packages/spyder_kernels/customize/spydercustomize.py", line 110, in execfile
exec(compile(f.read(), filename, 'exec'), namespace)

File "/home/zly/pyCFTrackers/eval/eval_VOT2016.py", line 68, in
trackers), desc='eval eao', total=len(trackers), ncols=100):

File "/home/zly/anaconda3/lib/python3.7/site-packages/tqdm/_tqdm.py", line 1022, in iter
for obj in iterable:

File "/home/zly/anaconda3/lib/python3.7/multiprocessing/pool.py", line 748, in next
raise value

ValueError: max() arg is an empty sequence

i want to run the cf_demo.py to detect my own dataset, but how to get the groundtruth_rect.txt ?

你好，我在pycharm上运行了您的代码，可是一直报错编译不成功。不知道是不是配置环境的问题。可以的话希望您可以出一个如何运行您文件的说明，这样会容易上手一些。非常感谢。

DSST ；Staple；CSRDCF；SAMF； DSST-LP；MKCFup；MKCFup-LP；MCCTH-Staple；BACF；运行时，都会报错：

Segmentation fault: 11

Stack trace:
[bt] (0) /home/mo/.conda/envs/tracker-py36/lib/python3.6/site-packages/mxnet/libmxnet.so(+0x40c1900) [0x7f1a0411e900]
[bt] (1) /lib/x86_64-linux-gnu/libc.so.6(+0x3efd0) [0x7f1a5fa94fd0]
[bt] (2) /home/mo/pyCFTrackers/lib/eco/features/_gradient.cpython-36m-x86_64-linux-gnu.so(grad1(float*, float*, float*, int, int, int)+0x96) [0x7f19e00069d6]
[bt] (3) /home/mo/pyCFTrackers/lib/eco/features/_gradient.cpython-36m-x86_64-linux-gnu.so(gradMag(float*, float*, float*, int, int, int, bool)+0x1a1) [0x7f19e00070e1]
[bt] (4) /home/mo/pyCFTrackers/lib/eco/features/_gradient.cpython-36m-x86_64-linux-gnu.so(+0x15e7) [0x7f19e00065e7]
[bt] (5) /home/mo/.conda/envs/tracker-py36/bin/../lib/libpython3.6m.so.1.0(_PyCFunction_FastCallDict+0x229) [0x7f1a60ad5a29]
[bt] (6) /home/mo/.conda/envs/tracker-py36/bin/../lib/libpython3.6m.so.1.0(+0x147f4c) [0x7f1a60b5af4c]
[bt] (7) /home/mo/.conda/envs/tracker-py36/bin/../lib/libpython3.6m.so.1.0(_PyEval_EvalFrameDefault+0x26fd) [0x7f1a60b5dbbd]
[bt] (8) /home/mo/.conda/envs/tracker-py36/bin/../lib/libpython3.6m.so.1.0(+0x1464c0) [0x7f1a60b594c0]

ERROR: mxnet-cu90 1.4.1 has requirement numpy<1.15.0,>=1.8.2, but you'll have numpy 1.16.2 which is incompatible.

i can't import the '_gradient' by following steps. how can i solve this?

I have the same problem about running this on WIN10...

如何将otb100 数据导入进去 自己创建一个dataset文件夹吗

ImportError: cannot import name 'gradient' from 'lib.eco.features' (E:\PyCharmProjects\KCF-master\lib\eco\features_init.py)
在\lib\eco\features下的_gradient只有一个.cpp文件，这个import导入出错，是啥原因呢？求大神指点迷津！

there is no data to even run an example. your description in readme is incorrent. python cf_demo.py not working. and in repo no folder dataset, and also no any documentation about format or content of this directory. wtf?

PS C:\Users\username\Desktop\pyCFTrackers\examples> python .\cf_demo.py Traceback (most recent call last): File ".\cf_demo.py", line 7, in <module> data_names=sorted(os.listdir(data_dir)) FileNotFoundError: [WinError 3] Системе не удается найти указанный путь: '../dataset/test' PS C:\Users\username\Desktop\pyCFTrackers\examples>

Hi! i tried your trackers on video and they work very slow for me (~10 second on iteration). i'm tried to use moose and kcf trackers.
i'm use:
cupy-cuda100 7.1.1
mxnet-cu100 1.5.1.post0
Cuda compilation tools, release 10.0, V10.0.130
what could be the problem?

Can you share the file structure of your codebase, i.e locations of datasets wrt cf_demo.py.

您好！感谢您的分享。我想请问下您里面集成的这些模型是不是都是纯手工特征不需要训练的呢？我没找到任何的模型文件。期待您的回复。

How can I get the test data in demo?

It seems that your OTB100 result of ECO is lower than it paper and code from https://github.com/StrangerZhang/pyECO, do you have any idea about how to fix this?

I got a success of 0.683 and precision of 0.911 using code from pyECO.

请问您在使用pysot-toolkit的时候很顺利吗，我遇到很多问题，并不能正常的使用这个工具。主要问题是region不能识别和numba报错，想知道您是如何使用的