2019-04-25 11:24:16 -0600 | marked best answer | CSRT tracker psr_threshold meaning and usage Hi all, I understand that using a different value for psr_threshold in CSRT tracker affects the reliability of the tracker. We made some tests and found out, that the default value of 0.035 is too small for us. Is there any recommended way of changing this parameter instead of just wild guessing based on empirical testing? Can somebody please explain the meaning of this parameter? What is a reasonable minimum or maximum value for psr_threshold ? KCF tracker, for example, has the params::detect_thresh value which seems to be a percent reliability value so it's easy to use. Thanks. Alex |
2019-04-25 11:24:03 -0600 | commented answer | CSRT tracker psr_threshold meaning and usage Thanks for all the valueable info. A. |
2019-04-25 04:29:02 -0600 | received badge | ● Student (source) |
2019-04-25 02:45:15 -0600 | received badge | ● Enthusiast |
2019-04-24 08:27:10 -0600 | asked a question | CSRT tracker psr_threshold meaning and usage CSRT tracker psr_threshold meaning and usage Hi all, I understand that using a different value for psr_threshold in CSR |
2019-04-17 10:16:34 -0600 | marked best answer | Tracking API with data in GPU memory Hi all, looking at the tracking API I was wondering if the init or update methods can be fed with GpuMat data instead of Mat. I.e. if I have decoded data in GPU memory, can I use the tracking API directly or should I copy data from GPU to host first? Thanks. Alex |
2019-04-17 10:16:34 -0600 | received badge | ● Scholar (source) |
2019-04-17 10:16:26 -0600 | commented answer | Tracking API with data in GPU memory Thanks, that's what I was thinking. There are some interesting projects like this one which show that the tracking algos |
2019-04-16 10:25:47 -0600 | asked a question | Tracking API with data in GPU memory Tracking API with data in GPU memory Hi all, looking at the tracking API I was wondering if the init or update methods |
2018-10-20 03:43:05 -0600 | commented answer | How to run OpenCV DNN on NVidia GPU I see, thanks. But to come back to the original question, because I am still not clear about it. Does that mean that we |
2018-10-19 05:41:57 -0600 | commented answer | How to run OpenCV DNN on NVidia GPU Hm, that's a bit surprising though, because OpenCV used to inform about and already has CUDA support ( https://docs.open |
2018-10-19 05:01:59 -0600 | commented answer | How to run OpenCV DNN on NVidia GPU Ok, does that mean that Yolov3 (which has been added to OpenCV) cannot use cuDNN for maximum speed? If not, are there pl |