"NOTE we not use gmm to generate low, high, peak value" ------in pysot-oolkit/pysot/evaluation/eao_benchmark.py
The result evaluated by the tookit is not so correct, because the values of "low"、“high "、"peak" is fixed, but it should be auto computed by KDE(GMM) , I will appreciate it if the author can update the compute code.
"NOTE we not use gmm to generate low, high, peak value" ------in pysot-oolkit/pysot/evaluation/eao_benchmark.py
The result evaluated by the tookit is not so correct, because the values of "low"、“high "、"peak" is fixed, but it should be auto computed by KDE(GMM) , I will appreciate it if the author can update the compute code.