py-motmetrics
py-motmetrics copied to clipboard
cheind
:bar_chart: Benchmark multiple object trackers (MOT) in Python
I used py-motetrics because I wanted to get mota, motp for the custom dataset, and I got the results. Mota outputs a value around 0.8, while motp outputs a value...
Hello @cheind and @jvlmdr, You've both done an excellent job in clarifying various issues. It really helped me use this work better! Just a quick question - Consider a ground-truth...
Description: I am trying to evaluate the performance of a multi-camera multi-target tracking system and I would like to use the IDF1 metric. However, I am not sure how to...
In `MOTAccumulator.merge_event_dataframes` （Line 443), we use `copy.index.map` to update index. When `copy` is an empty DataFrame, the size of `copy.index` is 0. In this situation, `copy.index.map` will return an object...
Running [this](https://github.com/cheind/py-motmetrics#for-custom-dataset) example gives different results than running `motmetrics.apps.eval_motchallenge`: ``` IDF1 MOTA acc 0.71134 0.552 ``` vs ``` IDF1 MOTA acc 73.2 53.7 ``` Their default configuration seems to be...
idp and idr value is same, also idp and precision or idr and recall value is same too. I don't understand this issue. I used this code snippet: ``` track_lines...
Currently I am searching more about this metrics. Using the IoU distance we obtain a value for MOTP, for example 0.8. What that means? What is the range for the...
This PR resolve #190 issue. When we utilize motmetrics with other packages such as nuscenes-devkit, options = {"ana": ana} sometimes cause errors. This PR set options = {} when ana...
